Number 2 /2024

INTEGRATION OF THE SURICATA INTRUSION DETECTION SYSTEM AND OF THE WAZUH SECURITY INFORMATION AND EVENT MANAGEMENT FOR REAL-TIME DENIAL-OF-SERVICE AND DATA TAMPERING DETECTION AND ALERTING

Gheorghe-Romeo ANDREICA, Iustin-Alexandru IVANCIU, Daniel ZINCA, Virgil DOBROTA

Abstract: This paper addresses one of the cybersecurity challenges posed by the rapid growth of IoT and intelligent transport systems. It aims to develop a security monitoring and alerting system for GPS devices in these systems, integrating the Suricata Intrusion Detection System (IDS) mechanism and the Wazuh Security Information and Event Management (SIEM). The solution is focused on detecting, alerting and real-time monitoring for Denial-of-Service (DoS) and Data Tampering attacks, ensuring robust protection against emerging cyber threats in IoT GPS tracking systems.

Keywords: Data Tampering, DoS, IDS, IoT, GPS Tracking, SIEM

DESIGN AND IMPLEMENTATION OF A FULL-DUPLEX GROUND STATION FOR THE QO-100 SATELLITE SYSTEM BASED ON SDR AND RASPBERRY PI

Nicolae CRIȘAN

Abstract: This paper proposes an approach based on the Adalm-Pluto SDR and Raspberry PI to design a ground station for geostationary satellite communications. The system is designed from modules to be versatile, agile, and scalable. In this case, the usage of modules is mandatory due to the complexity of the communications system, which simultaneously operates two radio channels at 2.4 GHz for uplink and 10 GHz for downlink. Such a combination with SDR (Software Defined Radio) and classic down and up converters benefits from using a Raspberry PI 4 with a powerful processor running under Linux. The proposed system allows the usage of GNU Radio with its powerful graphical interface written in Python. User interface UI has been undertaken by the Raspberry PI’s most versatile display with a touch screen of 7 inches. Another topic the paper addresses is using the helical antenna in front of the Horn antenna at the same focus point established by the dish reflector. The ground station proposed here has been successfully tested to transmit short radio packages containing data about temperature, atmospheric conditions, and voice.

Keywords: Software-defined radio, QO-100, Geostationary orbit (GEO), Right-handed circularly polarized (RHCP)

KUBERNETES CONFIGURATION FILES CREATED WITH A PYTHON QUART WEB INTERFACE FOR REAL LIFE SCENARIOS OF KUBERNETES DEPLOYMENTS

Ioan-Alexandru DUMITRE, Szabolcs-Gavril RUGINA, Laszlo-Csongor LENGYEL, Andrei-Bogdan RUS, Virgil DOBROTA

Abstract: This paper presents the development of a web interface using Python Quart to simplify the creation and management of Kubernetes manifest files. The interface allows users to create, update, delete, and manage Kubernetes resources through an intuitive platform. Developed using Test-Driven Development (TDD) principles, the application minimizes errors and improves reliability by ensuring that each component is rigorously tested before deployment. By leveraging Quart’s asynchronous capabilities, it handles multiple requests efficiently, enhancing scalability and performance. Performance tests with Locust and real-time monitoring with Prometheus show the system's ability to manage multiple users and highlight areas for improvement in request latency.

Keywords: Asynchronous Programming, Docker, Kubernetes, Python Quart, Test-Driven Development, WSL

Number 1 /2024

CHATBOT DESIGNED FOR INTERNATIONAL STUDENTS

Carla-Mihaela BULZESCU, Lacrimioara GRAMA

Abstract: The main purpose of this paper is to present a chatbot developed for international students. It provides guidance on a range of challenges, including safety, culture shock, housing, language, and academic notes. For this purpose, a dataset of 72 question-and-answer intents was developed from scratch, designed to assist students. The training phase uses the dataset to create the neural network model. The chatbot makes use of established rules and techniques for natural language processing to predict the appropriate responses to user’s demand. Two Python-based versions of the chatbot are developed: one with a user interface, designed using a Flask server connected to a front-end section, and one with a compiler-based method that can handle speech input and output. Both versions achieved high accuracy during the training phase, suggesting good forecasting for the purpose upon which the chatbot was designed.

Keywords: chatbot, artificial intelligence, neural network, natural language processing, text to speech

AN ANALYSIS OF THE PERFORMANCE OF DIFFERENT DPSK MODULATION DATA FORMATS USING CROSS-POLARIZATION MODULATION BASED 80 Gb/s ALL-OPTICAL WAVELENGTH CONVERSION IN A SINGLE WIDEBAND SOA

SAROJINI R , SIVANANTHA Raja A, SELVENDRAN S

Abstract: At key optical network nodes the efficient deployment of WDM/DWDM technologies can be made possible by wavelength conversion.  In this research, Cross-Polarization Modulation (XpolM) based All-Optical Wavelength Conversion (AOWC) using wide band Semiconductor Optical Amplifier (SOA) is obtained. The effectiveness of various optical differential phase modulations on wavelength conversion is analysed. NRZ-DPSK/DPSK, 33%RZ-DPSK, 50%RZ-DPSK and 67%RZ-DPSK/CSRZ-DPSK are investigated here. The data rate is 80 Gb/s and the conversion bandwidth is 1.04 nm. The analysis is extended for Linear, Lorentzian and No-approximation material gain simulation models. The NRZ-DPSK performs better in terms of maximum coupled intensity and ellipticity. CSRZ-displays a power spectral gain of up to 12 dB and shows narrower spectral width.

Keywords: Semiconductor Optical Amplifier (SOA), Cross-Polarization Modulation (XpolM), All-Optical Wavelength Conversion (AOWC), Differential Phase Shift Keying (DPSK), Carrier Suppressed Return-to-Zero DPSK (CSRZ-DPSK)

INTRODUCTION TO CYCLIC CODES AND ILLUSTRATION OF THEIR IMPLEMENTATION THROUGH LINEAR FEEDBACK SHIFT REGISTERS WITHIN A PROGRESSIVE WEB APPLICATION

Ioana Lavinia SOFIA

Abstract: This paper provides an introductory overview of cyclic codes of Hamming type and showcases their effectiveness in detecting and correcting errors in digital communications. Thus, a Progressive Web Application has been developed through React.js and p5.js JavaScript libraries to simulate the coding operations using Linear Feedback Shift Registers. This GUI serves an educational purpose, being used in Information Transmission Theory laboratories so that students could easier comprehend this particular coding scheme. The results obtained highlight the efficiency of developed software application in the e-learning process of cyclic code of Hamming type and its adeptness in error detection and correction within digital communications.

Keywords: cyclic codes, digital communications, LFSR, React.js, p5.js, Progressive Web Application, GUI

EXTENDED LOCAL BINARY PATTERNS-BASED POLARIMETRIC SYNTHETIC APERTURE RADAR IMAGE CLASSIFICATION

Andreia Valentina MICLEA-CECALACA, Ioana Ilea, Romulus TEREBES

Abstract: Polarimetric Synthetic Aperture Radar (PolSAR) imaging has emerged as a powerful tool for remote sensing applications, providing valuable insights into the physical properties of observed targets. Among various methods employed for PolSAR image analysis, the Local Binary Patterns (LBP) algorithm has gained significant attention due to its robustness and efficiency in texture classification tasks. However, conventional LBP approaches often encounter challenges in capturing complex texture patterns and preserving discriminative information inherent in PolSAR images. In this paper, we propose an integration of the classical Extended Local Binary Patterns (ELBP) method into a framework specifically designed for PolSAR image analysis. The ELBP method integrates advanced feature extraction techniques based on multiresolution analysis to enhance the discriminative power of texture descriptors. By exploiting the polarimetric information embedded in PolSAR images, the ELBP method effectively captures intricate texture characteristics while preserving spatial context. Comparative analyses against state-of-the-art texture analysis techniques demonstrate the efficiency of the ELBP method in terms of classification accuracy.

Keywords: ELBP, PolSAR imaging, polarization, SVM

WEBRTC AND PJSIP IN ASTERISK: A CLOUD-BASED APPROACH

Denisa-Oana SELIN, Alin-Tudor SFERLE, Tudor-Mihai BLAGA, Daniel ZINCA, Virgil DOBROTA

Abstract: This paper presents our design of launching WebRTC and PJSIP in an IP-based Private Branch Exchange instance running in OpenStack private cloud. This implementation was done using Ubuntu 20.04 LTS-based Asterisk in a Triangle topology (browser-server-browser). The major contributions were the following: (1) design how to integrate open-source software devices working with WebRTC in a private cloud; (2) security and encryption ensured by generating certificates and setting them appropriately in the terminals; (3) metrics monitoring along with call logs available directly in the browser by integrating Asterisk with Grafana Cloud. We had three scenarios: (1) PJSIP-to-PJSIP; (2) WebRTC-to-WebRTC; (3) PJSIP-to-WebRTC.

Keywords: PJSIP; private cloud; WebRTC

Number 2 /2023

UHF-RFID ANTENNA FOR SEMI-ACTIVE ASSISTED TAGS

Nicolae CRIȘAN

Abstract: This paper proposes a new concept of using RFID tag antennas more actively in the identification process. This innovative concept is based on antenna native properties that allow RF harvesting for DC biasing, either from the reader or from other proximity RF sources at the same time. Nevertheless, for now, RFID identification tags without a self-DC power supply have become more popular and are referred to as passive. Eventually, both RF and light harvesting processes could be addressed as a foreseeable solution for identification range extension. Soon this trend will lead to the cheapest (printable) RFID tags at competitive prices closely shifted to those of the already mass-produced barcode tags, mainly in use today. The proposed semi-passive tag antenna is working by replacing the IMPINJ-MONZA chip with the EM4325 chip which can be DC supplied by a PV panel. This tag can work as RF energy harvester and can be DC biased as well, being the most capable for the proposed concept. The antenna is printed on paper using a metal powder ink printer and the tag is interrogated using a UHF-RFID reader. The harvesting for tag DC biasing is extended through a very small solar panel, which acts also as a substrate for the RFID antenna. Thus, the extended concept of the semi-active tag could increase the coverage and the sensing range of the reader significantly. This happens especially in daylight or when other nearby RF sources are available, otherwise, the tag will work as an ordinary one. This approach will allow the antenna to be shrunk even further near the Chu-Harrington limit.

Keywords: Antenna, RFID-tag, radio frequency identification, internet of green things

SECURITY ISSUES IN INTERNET OF THINGS BOTNETS: A HIGH INTERACTION HONEYPOT APPROACH

Alexandru LAZAR, Tudor-Mihai BLAGA, Daniel ZINCA, Virgil DOBROTA

Abstract: For a better awareness of the tactics employed by the malicious entities in Internet of Things, a system called Honeypot tricks the attackers into exploiting its “sweet” fake resources. While implementing nine types of devices only six of those were attacked. Despite this, the honeypot managed to capture attacks destined for three devices that were not implemented. Overall, several real-world attacks were captured and analyzed providing different indicators of compromise. No new threats were identified but the server only ran for a short period of time and with limited resources. This approach looks like a promising avenue for creating attacker profiles and collecting data on botnet behavior.

Keywords:  botnet; exploit; honeypot; IoT

MOBILE APPLICATION FOR BLE INDOOR POSITIONING

Raul MĂLUȚAN, Bianca BRISC

Abstract: Distance estimation in indoor localization is mainly solved by Received Signal Strength Indicator (RSSI). In our approach, the RSSI is determined empirically, and it’s used to estimate the distance between the user and beacons in a mobile app solution. Trilateration and Kalman filter help obtain better accuracy. The paper proposes a classical solution for indoor localization, but with features like blueprint implementation, Google Maps integration, and manual beacon positioning.

Keywords: localization, Bluetooth low energy, trilateration, beacon

Number 1 /2023

A NUMERICAL MODEL FOR I/Q MODULATION/DEMODULATION FOR AN FPGA-ENABLED SDR PLATFORM

Bianca L.M. BOTA, Rareș BUTA, Botond S. KIREI, Calin A. FĂRCAȘ, Sorin A. HINTEA

Abstract: This paper presents a numerical model for I/Q modulation/demodulation implemented in Matlab. This numerical model serves as a starting point for the HDL/FPGA implementation of a baseband processor for a previously developed SDR platform. The numerical model verification is carried out by performing 4, 8, 64, and 256-point QAM transmission over an additive White Gaussian Noise Channel and by comparing numerical and theoretical bit error rates. A partial HDL implementation of the baseband processor is given: the baseband processor generates a 1MHz tone signal that is mixed by the RF fronted, generating a double sideband modulated signal.

Keywords: Matlab, SDR, IQ Modulator, BER, QAM

DDoS ATTACK DETECTION USING SUPERVISED MACHINE LEARNING ALGORITHMS OVER THE CIDDOS2019 DATASET

Daniel ZINCA, Virgil DOBROTA

Abstract: Distributed Denial-of-Service (DDoS) attacks are one of the most common types of cyber-attacks that can cause severe damage to networks and systems. Traditional methods to detect them rely on signature-based Intrusion Detection Systems (IDS), which are limited by the need of prior knowledge of specific patterns and by the usual ineffectiveness against zero-day attacks. However machine learning (ML) algorithms have the potential to support the detection of new and unknown attacks. This article compares the DDoS detection performance of three Machine Learning techniques: Gaussian Naïve Bayes, Logistic Regression and Random Forest, based on validation metrics such as precision, recall and F1 score. The system was trained using three datasets extracted from CICDDoS2019 database. The results proved the detection of attacks at Layer 4 (TCP SYN/ UDP flood), and at reflective Layer 7 (MSSQL, NetBIOS). The Random Forests and Logistic Regression methods achieved a precision between 93.7% and 99.4 % over these three datasets.

Keywords  CICDDoS2019, DDoS, Gaussian Naïve Bayes, Intrusion Detection Systems, Logistic Regression, Random Forest

SECURE ACCESS WITH TELTONIKA GPS TRACKING DEVICES FOR INTELLIGENT TRANSPORTATION SYSTEMS

Gheorghe-Romeo ANDREICA, Ciprian STANGU, Iustin-Alexandru IVANCIU, Daniel ZINCA, Virgil DOBROTA

Abstract: GPS tracking devices are widely used in industries like logistics, transportation, and security. However, they are susceptible to cyber-attacks, including Man-in-the-Middle (MITM). This study focuses on Teltonika GPS tracking devices and examines the impact of MITM attacks on their operation. We propose implementing encryption protocols and other measures to enhance the security and resilience of Teltonika GPS tracking devices.

Keywords: GPS Tracking, MITM Attack, TLS/SSL encryption

SPATIAL-SPECTRAL CLASSIFICATION OF HYPERSPECTRAL DATA WITH CONTROLLED DATA SEPARATION

Andreia Valentina MICLEA, Mihaela ABRUDAN

Abstract: Exploring spatial-spectral data frequently involves classifying hyperspectral images using convolutional neural networks. Due to the high complexity of the data and the scarcity of available training samples, hyperspectral image classification presents significant difficulties. In the context of supervised classification, we find that traditional experimental designs are frequently misused in the spectral-spatial processing context, resulting in unfair or biased performance evaluation, particularly when training and testing samples are selected at random from the same dataset. Under these circumstances, the dependence caused by the overlap between training and testing samples may be artificially increased, in breach of the data independence assumption upheld by supervised learning theory. In order to prevent an unbiased classification result, we present in this paper a controlled strategy designed to minimize the overlap between the samples present in the training and the testing data sets. The proposed controlled sampling strategy ensures a more trustworthy generalization of the CNN model by minimizing the issues present in the random sampling approach, such as the inability to determine whether or not an increase in classification accuracy is due to the spatial information incorporated into a classifier or to an increase in the overlap between training and testing data sets. Experiments performed with a wavelet CNN on different HSIs, namely Indian Pines, Pavia University, and Salinas, ensure the generalization of the data under the assumption that the training and data sets are independent from one another, based on a controlled strategy. Considering the high dimension of the HSI image, as a pre-processing step, the evaluation of the proposed framework is done by PCA and FA methods.

Keywords: Hyperspectral images, classification, CNN, sampling strategy, hyperspectral image classification, spatial-spectral features, Wavelet

Number 2/2022

LEVERAGING SDN/SD-WAN IN VIRTUAL AND IOT DEPLOYMENTS THROUGH AUTOMATION

Robert BOTEZ, Gabriel LAZAR, Iustin-Alexandru IVANCIU, Virgil DOBROTA

Abstract: This paper proposes a solution for the automation of Software-Defined Networking (SDN)/ Software-Defined Wide Area Network (SD-WAN) testbeds using Infrastructure-as-Code and Configuration Management techniques. Through automation, our solution also aims to reduce the difficulties raised by the implementation of the testbeds in various topologies with several hardware architectures. A virtualized scenario including hosts, switches and an SDN Controller was leveraged by Vagrant and configured using Ansible. We devised a second topology for edge deployments, incorporating multiple Raspberry Pi boards, to validate the interoperability of the solution. The experimental results showed that the highest value of the configuration time represents no more than half of the time needed in case of a manual approach.

Keywords: Ansible, IoT, ONOS, SDN, SD-WAN, Vagrant.

DEEP LEARNING STRATEGIES BASED ON FINE-TUNING. APPLICATION TO MEDICAL IMAGE CLASSIFICATION

Stefania BARBURICEANU, Andreia MICLEA, Romulus TEREBES

Abstract: This research focuses on the use of fine-tuning techniques to analyze various scenarios where some of the learnable layers in a CNN model are frozen and some are fine-tuned. The experimental section involves popular CNN architectures such as AlexNet and Vgg, and analyzes the classification performance of such models by freezing successively the convolutional and fully connected layers. The classification layers in the proposed technique are substituted with new ones that are trained starting with random values, while the other layers are either frozen or fine-tuned. For the frozen layers, the knowledge obtained from training on a vast dataset is applied to the new classification task, whereas for fine-tuned layers, the network parameters are changed to suit the new problem. The goal of this research is to identify the network location that enables the ideal balance between the generality and specificity of extracted features. The experimental study was undertaken using a small public dataset of medical images. The results are also compared against those obtained when fine-tuning is performed for all layers, as well as when all network parameters are directly transferred from pre-trained models for various learning rate values. For the considered models, the point that offered the optimum compromise between specificity and generality was found. The worst results are obtained when no layers are frozen. When utilizing the same weights from the pre-trained model, the results demonstrate that the classification scores may be improved by appropriately modifying the learning rate of the new classification layers.

Keywords: IC verification, adaptive sampling, optimizer

MAPPING THE VIRTUAL EDUCATIONAL TOURS FROM AN ECOSYSTEM PERSPECTIVE

Miruna CONŢ , Aurelia CIUPE

Abstract: Recent studies related to Distributed Denial-of-Service (DDoS) attacks were focused on determining how to leverage data plane programmability, enabled by P4 language, to identify intrusions directly in network switches, without the involvement of Software-Defined Networking (SDN) controllers. This facilitates the detection of TCP SYN flooding, for instance, without changing the flow paths through a central controller, therefore increasing the speed for managing workloads. In this paper, we propose a novel mechanism by introducing the capability of P4-based data planes to implement real-time detection and to respond effectively to attacks that require stateful functionalities (e.g., port scan).

Keywords: virtual tour, virtual environment, virtual world, education, exploration techniques, students' opinion

A SYN FLOODING DDoS ATTACK DETECTION IN P4-BASED PROGRAMMABLE NETWORKS

Calin-Marian IURIAN, Daniel ZINCA, Iustin-Alexandru IVANCIU, Tudor-Mihai BLAGA, Virgil DOBROTA

Abstract: This paper presents a discrete time simulation procedure of switched mode power supply (SWPS) power stages using Euler’s forward method for numerical integration. Usually, the simulation of SMPS power stages is carried out in PSPICE alike circuit simulators, but sometimes this is inconvenient and slow. The proposed simulation method may be used for educational purposes, whenever a quick and easy illustration of the SMPSs is desired. Furthermore, it can be deployed in the development of digital control algorithm, as the method may be implemented in logic simulators (event driven or delta time simulators). This discrete time modeling procedure can be implemented in any number crunching environment (Matlab, Octave, NumPy, etc.), in a logic (event-driven) simulator or in a simple spreadsheet. The procedure consists in (i) writing the ordinary differential equations (ODEs) of the modelled power stage, (ii) solving the ODE using numerical methods. Several power stage models were developed and simulated: (i) an ideal buck-boost was simulated in MATLAB, (ii) a boost converter simulated in a spreadsheet and (iii) an ideal synchronous buck converter was described in a hardware description language, VHDL. The numerical error between discrete time and PSCIPE simulation results are negligible.

Keywords: DDoS, P4, SDN, TCP SYN flooding, TCP SYN port scan

DISCRETE-TIME SIMULATION OF SWITCHED MODE POWER SUPPLIES USING NUMERICAL INTEGRATION FOR EDUCATIONAL PURPOSES

Botond Sandor KIREI, Calin Adrian FARCAS, Cosmin CHIRA, Ionut-Alin ILIE, Marius NEAG

Abstract: This paper presents a discrete time simulation procedure of switched mode power supply (SWPS) power stages using Euler’s forward method for numerical integration. Usually, the simulation of SMPS power stages is carried out in PSPICE alike circuit simulators, but sometimes this is inconvenient and slow. The proposed simulation method may be used for educational purposes, whenever a quick and easy illustration of the SMPSs is desired. Furthermore, it can be deployed in the development of digital control algorithm, as the method may be implemented in logic simulators (event driven or delta time simulators). This discrete time modeling procedure can be implemented in any number crunching environment (Matlab, Octave, NumPy, etc.), in a logic (event-driven) simulator or in a simple spreadsheet. The procedure consists in (i) writing the ordinary differential equations (ODEs) of the modelled power stage, (ii) solving the ODE using numerical methods. Several power stage models were developed and simulated: (i) an ideal buck-boost was simulated in MATLAB, (ii) a boost converter simulated in a spreadsheet and (iii) an ideal synchronous buck converter was described in a hardware description language, VHDL. The numerical error between discrete time and PSCIPE simulation results are negligible.

Keywords: Ordinary differential equations, Logic simulation, Switching converters, Euler’s Method, VHDL

Number 1/2022

AUTOMATED RECYCLING ARM

Ciprian CIUCIU, Laura-Nicoleta IVANCIU

Abstract: The importance of waste recycling is no longer a secret to anyone. However, the recycling rate in the world is relatively low, people are not interested enough in this problem. This paper the prototype of an intelligent, autonomous system, using live object detection, capable of sorting waste, with a low price and good speed. The waste objects are identified by means of a neural network, implemented on a Raspberry Pi board. The robotic arm, which picks up and transports the detected object into the appropriate bin, is controlled by an Arduino Nano board. The prototype was tested for various types of waste and the encouraging results show that it can successfully be implemented at an industrial scale, as part of waste management solutions.

Keywords: recycling, robotic arm, waste, object detection

A SYN FLOODING DDoS ATTACK DETECTION IN P4-BASED PROGRAMMABLE NETWORKS

Calin-Marian IURIAN, Daniel ZINCA, Iustin-Alexandru IVANCIU, Tudor-Mihai BLAGA, Virgil DOBROTA

Abstract: Recent studies related to Distributed Denial-of-Service (DDoS) attacks were focused on determining how to leverage data plane programmability, enabled by P4 language, to identify intrusions directly in network switches, without the involvement of Software-Defined Networking (SDN) controllers. This facilitates the detection of TCP SYN flooding, for instance, without changing the flow paths through a central controller, therefore increasing the speed for managing workloads. In this paper, we propose a novel mechanism by introducing the capability of P4-based data planes to implement real-time detection and to respond effectively to attacks that require stateful functionalities (e.g., port scan).

Keywords: DDoS, P4, SDN, TCP SYN flooding, TCP SYN port scan.

ON COMBINING SEVERAL OPTIMIZATION ALGORITHMS FOR IMPROVING THE EFFICIENCY OF IC VERIFICATION PROCESS

Alecsandra RUSU, Ingrid KOVACS, Bianca CĂRBUNESCU, Marina ŢOPA, Andi BUZO, Georg PELZ, Emilian DAVID

Abstract: In the modeling of complex integrated circuits, the locations of the sampled data points are essential to the success of the verification process. Most of the adaptive sampling methods based on iterative algorithms, that use data acquired from previous iterations to guide future sample selection, are optimization algorithms. An essential consideration of every adaptive sampling method is the trade-off between exploration and exploitation, i.e., the capability of the method to cover the under-sampled parts of the design space, but also focusing on interesting regions. This paper proposes three combinations of existing methods with the purpose of minimizing this trade-off and efficiently characterizing the verification space. Their performance and efficiency were validated on 20 synthetic test functions.

Keywords: IC verification, adaptive sampling, optimizer

FEATURE EXTRACTION METHODS BASED ON DEEP-LEARNING APPROACHES. APPLICATION TO AUTOMATIC DIAGNOSIS OF BREAST CANCER

Stefania BARBURICEANU, Romulus TEREBES

Abstract: This paper proposes the use of several pre-trained CNN architectures (Vgg16, AlexNet, and Resnet50) for texture feature synthesis. The information gained from training the models on a very large dataset (ImageNet) is transferred to a new classification problem (where the dataset is rather small) on the premise of transfer learning. For the classification part, the Support Vector Machine is considered. Since in the previous two decades, cancer has become more widespread, and by the year 2020, breast cancer was considered the most common disease on the planet, we are interested in applying the proposed strategy for detecting breast cancer. Various imaging modalities have been developed to diagnose this type of disease, with mammography being the most often employed. However, it is the expert's histological analysis that determines the ultimate diagnosis. As a result, pathologists may benefit from the development of autonomous diagnostic techniques based on computer vision technologies. Thus, a public database with histological images of malignant and benign breast tissues is employed in the conducted experimental analysis. For the same dataset, the present methodology outperforms other classical machine learning methods as well as CNN approaches.

Keywords: medical image classification, deep-learning, breast cancer, disease classification

PERFORMANCE EVALUATION OF ELK STACK VERSUS GRAYLOG AS OPEN-SOURCE LOG MANAGEMENT TOOLS

Roxana-Daniela ŞUŞTIC, Alexandra MORARU, Andrei-Bogdan RUS, Virgil DOBROTA

Abstract: This paper proposes a comparison between two log management tools, namely ELK Stack and Graylog, used as proactive measures for keeping track of changes in a network environment. Several experiments were based on performance metrics such as response time, CPU and memory usage, as well as on testing the security, alerting, monitoring, and data visualization capabilities. Performance-wise, ELK was the best solution due to its stability, faster response time and reliability under stress conditions. Graylog, on the other hand, was an easier to maintain, fitting the needs of a centralized log management, being more suitable for security and alerting purposes.

Keywords: containerization; Docker; ELK Stack; Graylog; log management; virtualization

Number 2/2021

POWER STAGE AND FEEDBACK LOOP DESIGN OF A FLYBACK CONVERTER FOR POWERING AN ELECTRIC SCOOTER

Bogdan Ștefan BUCȘE, Vlad BANDE

Abstract: This document represents a study about designing a power stage and a feedback loop for a Flyback converter – a circuit that is used to charge a battery pack designated for an electric scooter. This converter is designed to supply a 19.5V DC voltage and a maximum of 3.33A at the output of the circuit. The circuit consists of an EMI filter, an RCD Snubber protection circuit, a current mode PWM controller and a feedback loop designed using the TL431 integrated circuit. Its main purpose is to convert a 230V AC voltage to a stable 19.5V DC voltage. The research uses both the mathematical and simulation apparatus to model the behavior of a flyback converter before the physical design procedure. The results obtained will be used as milestones for checking the correct behavior of the real circuit.

Keywords: flyback converter, power stage, CCM operation, DCM operation, yype II Compensation, TL431

DATA MONITORING AND ACQUISITION SYSTEM FOR THE I2C PROTOCOL

Mark Eduard GROSS, Dorin Marius PETREUȘ

Abstract: This paper represents a resume of the study about a data acquisition system for monitoring and debugging an I2C (Inter Integrated Circuit) communication. The system has the role of a sniffer: it does not interfere with the data communicated between the master and slave. It can be connected to only one I2C bus at the time and it sends the data using the USB (Universal Serial Bus) to the computer. The acquired data will be displayed using an interface created in LabVIEW, while also being saved in a separate file. This system consists of a microcontroller with integrated SPI (Serial Peripheral Interface) peripheral and a SPI to USB transceiver. The USB connection will also power the device. The system was only designed and tested for I2C communications up to 400kbit/s (Fast-mode) with a 7-bit address.

Keywords: data acquisition, communication protocol, I2C, LabVIEW

A PYTHON-BASED FRAMEWORK FOR ADVANCED RESEARCH AND DEVELOPMENT ON SPECTRUM SENSING FOR COGNITIVE RADIO

George GROSU, Romulus TEREBEȘ

Abstract: The work of this paper is addressing the issue of narrowband spectrum sensing (SS) applications for cognitive radio networks assuming the OFDM signal over noisy and various fading channels with focus on the development of an open-source simulation and development software. The paper also proposes a residual and recurrent convolutional Neural Network (NN)-based method for spectrum sensing. This method is compared with a baseline energy detection (ED) approach. Detection performance of the models is analysed given various Signal-to-Noise Ratio (SNR) values using three-base scenarios. First two scenarios correspond to an additive white gaussian noise (AWGN) channel respectively Rayleigh flat fading channel and the third one corresponds to the frequency-selective Rayleigh fading channel since the SNR value and fading can affect drastically the detection performance in terms of false positives/false negatives leading to erroneous estimators. The experimental results are analysed through the receiver operating characteristics (ROC) plot containing the curves for both ED models enhanced by the denoising methods. On average, the classic ED algorithm with dynamic threshold outperforms the NN-based model, especially in low SNR domains. The NN-based model trained on constrained, tailored dataset characteristics outperforms the classic ED model in scenarios described by the corresponding characteristics.

Keywords: cognitive radio, energy detection, neural network, time series, spectrum sensing, wavelet

AUTO DAKAR: A WEB APPLICATION FOR THE MANAGEMENT OF AN AUTOMOBILE REPAIR SHOP

David GRUIAN, Iustin-Alexandru IVANCIU

Abstract: When it comes to an automobile service, it is very important to ensure not only good communication with the customers but also a strong online presence. This paper presents AUTO DAKAR, a highly customizable and easy to use web application designed to provide online visibility for the company and communication methods between customers and the automobile service. The application was developed with Laravel using the Voyager package for backend management, a Heroku server for testing and a DigitalOcean server for production. The experimental results highlight the main advantages of AUTO DAKAR: flexibility, easy maintenance and cost efficiency.

Keywords: automobile service, DigitalOcean, Heroku, Laravel, MySQL, PHP, web application

DEVICE LEVEL PROGRAMMABILITY USING RESTCONF PROTOCOL: AN INTRODUCTORY APPROACH

Vlad-Andrei MONORANU, Tudor-Mihai BLAGA, Virgil DOBROTA

Abstract: This paper presents an implementation of a virtualized testbed emulated with GNS3 version 2.2.19, controlled by Postman API client and by Python3 scripts. A GNS3 VM (virtual machine) running Ubuntu acted as remote server for hosting the network devices images. The aim was to involve RESTCONF (Representational State Transfer Configuration Protocol) which uses HTTP requests, as an alternative solution for NETCONF. The scenario included one Cisco IOS-XE-based virtual router CSR1000v 16.6.7 with RESTCONF enabled. By using a GET request on a specified URL, a list of capabilities of the router could be displayed. In the first scenario different other types of requests (PUT, POST, PATCH, and DELETE) have been checked, too. Another two Cisco CSR1000v routers were added in the network topology and their manual configuration was approached in the second scenario. Following static IPs allocations the behavior of the RESTCONF configured router was tested. As proof of feasibility, we enabled OSPF (Open Shortest Path First) routing protocol on all routers, assisting the discovery of neighboring devices. The major outcome of this project was that we were able to get long lists of capabilities that can be split in individual items and reused later for automatic configuration.

Keywords: Cisco router; GNS3; RESTCONF

Number 1/2021

ACOUSTICAL EVALUATION METHOD OF A CLASSROOM OF THE
TECHNICAL UNIVERSITY OF CLUJ-NAPOCA

Călin FARCAS , Marina TOPA

Abstract: In this paper, a method of measuring and interpreting acoustic quality in a classroom is presented. The acoustics of the
classroom are described by the reverberation time and the objective acoustic parameters that derive from it. The measuring
points were chosen in accordance with ISO 3382-1, so that to obtain an accurate acoustic characterization of the listening area. At
each measurement point, the impulse response was measured and then critical distance, reverberation time, clarity and definition
were extracted. Finally, the values of the objective acoustic parameters were mapped and the results were analyzed.

Keywords: acoustic measurements, reverberation time, objective acoustic parameters, critical distance

STABILITY OF SEVERAL TYPES OF HIGH FREQUENCY RING OSCILLATORS DEPENDING ON CIRCUIT CONFIGURATIONS AND PARAMETERS

Călin FARCAS , Ramona AGRIJAN, Sorin HINTEA

Abstract: This paper proposes studying the performance of the PLL loop depending on the configurations of the voltage-controlled oscillators used and on the types of ring oscillators. The oscillation frequency depends on the oscillator structure, the size of the CMOS transistors, the supply voltage, the control voltage, and the load capacity. In several configurations, ring oscillators and VCOs were used, with several gate types and several sizes of CMOS transistors. We also added an inverter to the output of the studied oscillators and demonstrated the increase of the robustness of a PLL loop in a larger frequency range. By measurements, we have shown that the oscillation frequency is no longer influenced by the variation of the load capacity. We measured the phase-locked time for the PLL loops and demonstrated that by using the inverter at the oscillator's output to the load, we obtain the shortest time. The implementation and simulation of the circuits used in this paper were performed in the LT Spice XVII simulator. The simulations were performed transiently, with the duration set in such a way that the results are interpreted properly.

Keywords: ring oscillators, oscillation frequency, PLL loop, phase-locked time

SECURITY OPTIMIZATION OF IOT PLATFORMS BASED ON NAMED DATA NETWORKING

Dan-Andrei MARGIN, Denisa-Adina MOLDOVAN, Iustin-Alexandru IVANCIU,
Jordi DOMINGO-PASCUAL, Virgil DOBROTA

Abstract: When it comes to developing a smart system involving sensors and actuators, there are two main problems to be addressed: what platform to be used as infrastructure and how to develop the security layer? In this paper, Orion Context Broker (OCB) is chosen for data management, while security layer is composed by Named Data Networking (NDN) and FIWARE modules (IdM KeyRock and Wilma PEP Proxy). The work was focused on data confidentiality and integrity because this is an important aspect of an IoT system. Store capabilities of OCB were extended from one value for an attribute to long term store archive. This approach is benefic for obtaining data necessary for statistics and predictions, where historical data must be used.

Keywords: confidentiality, data archive, FIWARE, integrity, IoT devices, Named-Data Networking, security

PLANT DISEASE CLASSIFICATION USING TEXTURE AND COLOUR

Stefania BARBURICEANU, Romulus TEREBES

Abstract: Plant disease detection and classification is an important topic from the precision agriculture domain. Since diseases have a strong negative impact on agricultural production, they must be detected correctly and as soon as possible in order to be able to take the most appropriate measures. We propose a feature extraction technique which uses texture and colour for the classification of plant diseases. The textural content is extracted from grey-scale images, while the colour distribution in the images is generated from RGB samples. The two types of information are derived independently and are fused together through concatenation. For the experimental evaluation, a public database containing real images of healthy and non-healthy plant leaves was used. We considered 4 setups: corn, grape, potato and tomato disease classification. The colour information added by the proposed approach increases the performance compared to using only the textural information. Even if our technique is not able to surpass a joint texture-colour feature extraction method, it provides more efficiency from the computational time point of view. The proposed method is a good compromise between classification performance and computational complexity.

Keywords: image classification, plant disease detection, texture, colour

A NOVEL LOCAL BINARY PATTERNS AND WAVELET TRANSFORM-BASED APPROACH FOR HYPERSPECTRAL IMAGE CLASSIFICATION

Andreia Valentina MICLEA, Romulus TEREBES

Abstract: PIn this paper we propose a novel approach for the classification of hyperspectral images based on a parallel architecture. The model is constructed based on different representations of the hyperspectral data along the spatial and the spectral dimensions. On the spatial domain we use Local Binary patterns-based histograms whereas along the spectral dimension of the hyperspectral dataset Wavelet filters are employed. These two descriptors are concatenated in a unique feature vector and a Support Vector Machine classifier is further used for classification. We show in the experimental section that the proposed method improves the accuracy on standard datasets (Indian Pines and Salinas) when compared to similar designs that consider Fourier transform-based descriptors.

Keywords:SVM, hyperspectral, local binary patterns, wavelet, spatial-spectral information

Number 3-4/2019

BATTERY AS INDEPENDENT POWER SOURCE

Levente SZABO

Abstract: Batteries are charged with energy, they store this energy and give it back if needed. Basically it works through consecutive charging-discharging cycles with a subunit efficiency. Inside the batteries that are in a idle state (not charging and nor discharging) there are particles in motion with electric charge and also there can be reversible electrochemical reactions in both directions simultaneously. Their energy can be extracted under special conditions and used as electric energy. This offers the possibility of using electric batteries as independent sources of electricity. If we add a special electrical circuit to the battery, it will not be necessary to charge it from outside. It will be able to supply a certain amount of energy permanently.

Keywords: pulse charging, battery , independent power source

Continuous Operation Monitoring of Electronic Circuits with Embedded Microcontroller

Gyorgy GYOROK

Abstract: Nowadays, it is commonly used when you can query the state of an industrial equipment, appliance, household appliance, vehicle status, technical parameter, using a computer. In such cases, a certain algorithm running on a computer can be used to query certain archived logs via a built-in interface, or to diagnose the status parameter of a component, or to specify more precisely the errors. For electronic devices, the reliability of the circuits is only provided through the manufacturing technology used. Typically, a goodness-coefficient for an electronic circuit is not measured, and usually no exact method is used to detect the circuit state, the reliability of an electronic circuit. This article describes some details of a research where a microcontroller is used together with the supervised circuit for continuous state monitoring and diagnosis.

Keywords: embedded, microcontroller,state monitoring, diagnosis

Number 2/2019

THE MEDIATING EFFECT OF THE BRAND ON THE RELATIONSHIP BETWEEN SOCIAL NETWORK MARKETING AND CONSUMER BEHAVIOR

Sara HASANI, Mahdi BAHAGHIGHAT, Mehdi MIRFATAHIA

Abstract: : In this paper, we investigate the relationship between social network marketing and consumer behaviors. It is a descriptive study with primary focus on practical aspects. To do the task, we conduct a survey based on our proposed questionnaire. The Structural Equation Modelling (SEM) and Smart PLS software are used to evaluate the obtained dataset including more than 384 samples. The results indicate that the brand value has more impact on consumer response as compared to social networks marketing.

Keywords: social media marketing, consumer behavior, Telegram, Instagram, SEM, PLS.

DEDICATED APPLICATIONS OF TELEPRESENCE ROBOTS FOR EDUCATION

Paul TOTA, Mircea-F. VAIDA

Abstract: Some companies have launched telepresence robots with applications in education, management, conferences and workshops. Robots available in commerce are somewhat simple, with just one tablet, a smartphone, or a computer at the user. The robot itself is made up of a robot car, a telescope bar, a display (tablet, smartphone, or dedicated crystal display), software and it makes a realistic remote communication that enable the user to interact effectively with the remote environment. A challenge in the field of research and design of telepresence robots is the realization of humanoid robots that allow the human user to connect to the robot through wireless technologies and with sensors to take on environmental sensations and interact with the environment where the robot is. Until such robots become available to the general public, several steps should be taken: solving the balance problems and digital control of the current humanoid robots, customizing the public with robotic humanoid applications, studying the interaction between people and telepresence robots and using the data obtained in the design of robots. In the educational system, telepresence robots can make possible the virtual participation of the students with disabilities, that can’t be transported to school, to the class for taking part at the courses. Also, telepresence robots can be used for multimedia educational projects which involves filming and taking photos in dangerous areas for humans such as: caves, small spaces or areas with radiation.

The problem of the commercial telepresence robots, to be used in education, consists in two elements: they have big dimensions, weight (not easy to carry for kids) and are expensive. In this paper we present three projects as alternatives to commercial telepresence robots. The first two are based on commercial educational kits that are easy to mount, have software installed, but are less flexible for applications dedicated to telepresence robotics. The third project is made with open source components that offer a lot of flexibility at a low price and allows the implementation in the software of the desired mathematical model. We also have analyzed, in this paper, the interaction of secondary and high school students with the telepresence robots, so we have find some future directions of research like: intelligent autonomy dedicated software, a commercial design and the possibility

to be easily personalized for each user or class of students.

Keywords: telepresence, robot, Arduino, applications, education, television, operator, photography

NEOTERIC MITIGATION SCHEME FOR PILOT CONTAMINATION IN

TDD MASSIVE MIMO

Jignesh M. PARMAR

Abstract: A future generation wireless networks have rapidly grown that’s demand more power efficiency and spectral efficiency. With This Demand In The Current Scenario, Massive MIMO Have Capable Technology. A Massive MIMO Can Support The Aerials in A Huge Quantity On Each Base Station. For Avoid Interference within & between the cells, All User Should Be Allocated With An Orthogonal Pilot Sequence. Practically Pilot Sequences Are Limited in Number and Therefore as the Number of Users Increases, It’s Not Capable for Each User to Have a Different Pilot Sequence. So, The Same Pilot Sequences Are Reused In Different Cells Which Caused Inter-Cell Interference. This Phenomenon is known as ‘Pilot Contamination”. In This Work we have propose to mitigate pilot contamination by adjoining the soft pilot reuse and weighted graph coloring algorithm to get the maximum achievable rate in uplink massive MIMO system. Simulation results shows that proposed combine method is provide better achievable rate in uplink compare to existing soft pilot and weighted graph coloring pilot decontamination..

Keywords: 5G, Massive MIMO, Time Division Duplexing, Pilot Contamination, Soft Pilot Reuse

REAL-TIME WATER QUALITY MONITORING SYSTEM FOR RIVERS

A. Sujaya DASGUPTA, B. Madhukar .S. ZAMBARE, C. Nitin KULKARNI, Arvind D.SHALIGRAM

Abstract: Rivers are the major sources of water supply in any city. These rivers have witnessed severe degradation of water quality owing to heavy industrialization, agriculture, and anthropogenic activities. A consistent rise in pollution is reported due to disposal of industrial effluents, domestic sewage, and agricultural runoffs into the rivers. Since, rivers are the major freshwater sources to any city and the high levels of pollution can be health threatening. Therefore continuous monitoring and analysis of their quality status enable us to send a report to the concerned authorities and further suitable measures can be thereby implemented to reduce the pollution levels. Also, such a system can draw public attention to the water quality status and aid in creating public awareness, also facilitating remote monitoring of water sources, data storage, and analysis in a cloud system. This paper proposes an Internet of Things (IoT) based design for real-time surface water quality monitoring and notification system, measuring the basic water parameters such as pH, temperature, dissolved oxygen, electrical conductivity, total dissolved solids.

Keywords: Real-time water quality monitoring, Internet of Things (IoT), Mula-Mutha

Number 1/2019

EXPERIENCES FROM RUNNING A “HOME-BREW” LABORATORY COURSE AND SETTING UP A NEW STUDENT MEASUREMENT SESSION FOR THE EDUCATION OF SENSORICS

Mihály GÖRBE

Abstract: : This paper summarizes the experiences that accumulated during the 10 semesters of education of the course titled “Physics of Sensors”, and presents a new student experiment for the training at ultrasonic proximity sensors at our sensorics student laboratory. The described pitfalls, needs, misconceptions, ineffective experiments helped to improve this laboratory course. Following the principles used during the creation of the student laboratory experiments, the presented modifications and new setup are implemented in a manner suitable for re-creating and modifying by the inexperienced student or the recently graduated bachelor at hobby or at work.

Keywords: physics of sensors, sensorics, sensor education, ultrasonic sensors, HC-SR04

INDUSTRY 4.0 PILOT LABORATORY

Károly SZÉLL

Abstract: By introducing the concepts of the Internet of Things (IoT) and the Cyberphysical System (CPS), industrial automation is undergoing significant changes. These concepts are not entirely new, but in recent years they have become more and more penetrating into industrial automation, placing automation systems in a very different light. They support the latest trends, such as higher level interconnection of system components, cognitive automation, and information gathering and processing in cloud-based applications. The application of IoT's and CPS's ideas in industrial automation has led to the definition of Industry 4.0, where 4.0 refers to the fourth industrial revolution allowed by Internet technologies to create smart products, smart production and smart services. This paper introduces the industry 4.0 pilot laboratory at Obuda University, Alba Regia Technical Faculty.

Keywords: Industry 4.0, IoT, digitization, smart manufacturing, OPC UA

ADVANCED NAVIGATION OF AUTOMATED VEHICLES IN SMART MANUFACTURING

Péter UDVARDY, Károly SZÉLL

Abstract: Smart manufacturing or also known as Industry 4.0 is the current trend in automated manufacturing. It includes IoT, cloud technologies, super-flexible automation and cooperative human-robot interaction to name a few aspects. Utilization of Autonomous Guided Vehicles is crucial in successful smart manufacturing. The paper reviews the most widely applied navigation strategies and current research directions related to AGV control.

Keywords: smart manufacturing, intralogistics, AGV

BATTERY VOLTAGE RESPONSE TO PULSE CHARGING

Szabó LEVENTE

Abstract: Study of the voltage response of Ni-Cd, Ni-Mh, Lead-acid and Li-Ion accumulators to rising voltage, falling voltage, two load pulses and two pulse trains for specific purposes. The electrochemical system in the battery always oscillates around an equilibrium point. When this state is disturbed by the external environment (electric field), the system tends to return to its original state by amplifying the oscillations around this point. The four types of accumulators behave generally similar, but in particular there are small differences in their voltage response. The particle response to electric field change is manifested by producing an electromagnetic wave that is added to the battery charging or discharging pulses. This wave is completely attenuated after a few oscillations. If the loading pulse period coincides with the period of these oscillations, the oscillations are sustained. The energy of these oscillations can be captured and used. Based on the experimental results, a battery model is proposed that describes the transition periods.

Keywords: pulse charging, Lead-acid, Ni-Cd, Ni-Mh, Li-Ion.

Number 4/2018

A REVIEW OF ECG BASED BIOMETRIC SYSTEMS

Liliana IVANCIU, Paul FARAGO, Sorin HINTEA

Abstract: : In the past years many studies suggested the use of electrocardiogram (ECG) for biometrics. Its unique properties make it particularly interesting for both soft applications as a standalone authentication method and for hard applications as a security enhancement layer. Compared to existing biometric identification systems, ECG is difficult to counterfeit and also provides aliveness indication. This paper presents an overview of the methods used for ECG based biometric systems and discusses the applications and security concerns

Keywords: biometrics, ECG, fiducial methods, identification, non-fiducial methods

DECISION BALANCING: PROCESSES IMPLEMENTATION BASED ON PIR SENSOR PERFORMANCES

Aurelian COSTEA, Paul ŞCHIOPU, Marian VLĂDESCU

Abstract: Along with the technological evolution in the field of infrared sensors and the exponential growth of the complexity of the serviced systems having more and more these sensors as components, it becomes increasingly important to decide with which devices a particular solution will be implemented or with which devices a particular maintenance situation will be solved. More and more devices, components and equipment are available on the supplier market, that claim to have similar sets of functional parameters and reliability. It is increasingly difficult to have a well endorsed scientifically detailed image to correctly, reliably and safely chose between all these devices, components and equipment available nowadays.

Keywords: infrared detection, PIR component, wavelength, beamsplitter detector, micro pyramids, plasmonic, driving circuit.

EFFECT OF TEMPERATURE ON THE PARAMETERS OF A BULK HETEROJUNCTION ORGANIC SOLAR CELL

Ana BĂRAR, Marian VLĂDESCU, Paul SCHIOPU

Abstract: This paper presents a theoretical study of the effect of temperature on the parameters of a novel polymer blend based organic solar cell, with a bulk heterojunction architecture. The variation of the short-circuit current density, the open-circuit voltage, fill-factor and power conversion efficiency with temperature will be analyzed.

Keywords: organic semiconductor, organic solar cell, temperature, mobility

APPLIED BEHAVIOR ANALYSIS OF NORTON AND THEVENIN CIRCUITS

Dumitru CIOFLICA, Adrian TULBURE and Calin PETRASCU

Abstract: The electromagnetic field theory uses the term „state” of the field, while the electronics circuits theory operates with the term „operation mode/regime” of the circuit. In the proposed paper the authors analyze two transient regimes, that take place in the current operation of the power grids: a) reducing the magnetic field of excitation circuit at the synchronous machine, assimilated to the coil discharging on a no-load grid and b) current supplying on an electric line with load, assimilated to the capacitor discharging on the connected consumers.

Starting from the standard basic models, the mentioned systems are mathematically modeled and simulated in MATLAB and LTSpice in three variants: aperiodical, periodical and critical mode. At the end of the paper a comparative evolution of the state variables u(t), i(t) at the dipoles terminals in different transient regimes is presented. The waveforms were experimentally validated by measurements on the test bench.

Keywords: power supply system, modeling-simulation, transient regime, Norton and Thevenin schemes, practical validation

Number 3/2018

RISKCAM: A SMART VIDEO SURVEILLANCE APPLICATION DESIGN CONCEPT

Elena Roxana BUHUS, Lacrimioara GRAMA, Alexandra DOBRE

Abstract: Fire outbreak, theft, logging, and poaching in the wildlife areas or illegal trespassing in the high security outdoor protected properties are one of the topics that should to be controlled, every day. We present a design concept for a smart application that could be capable of monitoring remote, several events according with the desired configuration and report the detected risk via email alert. This approach can successfully become the next level in surveillance applications based on smart video wireless cameras, since it provides a configurable approach of the application using just one type of wireless camera rather than a dedicated smart camera for each event detection, decreasing the costs. We named our target application RiskCam. Several features were considered for it that imply fire detection, abnormal movement detection (e.g. theft, trespassing), facial authentication and notification via email for the detected alarm (e.g. “Fire present in zone 1!”). We defined and tested the concept in MATLAB, using a modular approach, except the facial recognition module that was developed in C++.

Keywords: video surveilance, face recognition, fire detection, smoke detection, smart application

AUTOMATIC CHARACTER RECOGNITION IN PORCELAIN WARE

David Álvarez SANCHEZ, Santiago Garrido BULON, Luis MORENO, Adriana BIRLUTIU, Manuella KADAR

Abstract: This paper presents an automatic serial number recognition system based on computed vision techniques and machine learning. The serial number is engraved on the backside of porcelain ware. The approach uses standard algorithms for digit segmentation and a deep learning approach for the recognition. The system is developed within a project in collaboration with one of the biggest porcelain ware producer in Europe.

Keywords: Deep Learning, Computer Vision, Digit Recognition, Porcelain

MICROCONTROLLER’S USING IN MONITORING THE MAINTENANCE SYSTEM OF FLEXIBLE MANUFACTURING LINES

Alexandru NITU, Gheorghe MARC

Abstract: The paper presents a way of improving the interventions of the maintenance department made on the continuous manufacturing lines. The main purpose is represented by decreasing the times of stationaries of the machines and eliminating the loss of production. Using this system, the spent time for each planned reparation will also be decreased

Keywords: microcontrollers, industrial communications, databases, electronic system for maintenance, management of defects.

SOLVING MULTIOBJECTIVE OPTIMIZATION PROBLEMS USING A NON-PARETO BASED FITNESS ASSIGNMENT METHOD

Corina ROTAR, Laszlo Barna IANTOVICS

Abstract: Evolutionary algorithms for multi-criteria optimization fall into two major categories: Pareto and non-Pareto based techniques. Pareto techniques explicitly make use of the Pareto optimality concept in the assessment of individuals’ quality. In this regard, the extracting of the non-dominated solutions is a costly procedure in terms of time, but with remarkable results. Non- Pareto techniques avoid the drawback of the Pareto ranking techniques but having weaker performance. We believe that the non- Pareto fitness assignment in evolutionary multi-criteria optimization may be advanced. Therefore, the paper proposes a new procedure for fitness evaluation that accompanies a simple evolutionary algorithm for solving multiobjective optimization. The proposed evaluation procedure determines the potential of each solution to be non-dominated and does not identify the nondominated solutions, thus, the algorithmic efficiency in terms of time becomes superior to the Pareto-ranking based algorithms. The performance of a solution is given by the degree it may contribute to the improvement of the population and does not rely on its Pareto rank. Comparative analysis shows that the proposed evolutionary algorithm has lower complexity and provides better

results than the state of the art algorithm.

Keywords: Evolutionary Computation, multiobjective optimization, Fitness assignment

Number 2/2018

THE OPTIMIZATION OF TRAFFIC IN URBAN AGGLOMERATIONS

M. GAVRILĂ, I. BOGDAN

Abstract: Traffic jams are very frequent inside crowded cities due to the large number of vehicles. Improving the existing infrastructure entails prohibitive costs, which is why maximizing its use is very desirable. This paper presents a model for the optimization of traffic via dynamic modification of traffic routes in order to minimize the travel time up to the destination. Using dedicated communication capabilities, vehicles can communicate with the infrastructure in order to get updates on the status of the roads or any traffic jams occurring on one of the road segments along the configured traffic route. In a speed-crazed world and using an infrastructure (streets, highways) that failed to keep up in its development – and considering human error – traffic accidents and jams are imminent.

Keywords: Intelligent Transportation System (ITS), communication protocols, traffic simulator, European Commission, vehicles.

EXPERIMENTAL SIMULATION OF COHERENT 40 GBPS 2-DPSK DWDM LONG-HAUL FIBER OPTICAL SYSTEM WITH COUNTER-DIRECTIONAL EDFA

Tomáš HUSZANÍK, Ján TURÁN, Ľuboš OVSENÍK

Abstract: High speed and high capacity optical communication systems have become essential for todays high demand. High demand for broadband services and internet applications have been driving the development of optical networks rapidly. Transmission distance of fiber optical communication system can be further increased by incorporating fully optical amplification in the transmission route. The leading technology in optical amplification is based on erbium doped fiber amplifier (EDFA). In this paper we propose the 16 channel Dense Wavelength Division Multiplexed (DWDM) fiber optical communication system with transfer rate of 40 Gbps with counter-directional EDFA in program environment OptSim.

Keywords: Bit Error Rate, Dense Wavelenght Division Multiplexing, Erbium Doped Fiber Amplifier, Wavelength Division Multiplexing

A HIGH GAIN DUAL ISM BAND MONOPOLE PRINTED ANTENNA FOR WLAN APPLICATIONS WITH OMNIDIRECTIONAL RADIATION PATTERN

Priyabrata BISWAS, Srija DE, Debasree CHANDA SARKAR, Sushanta BISWAS, Partha PRATIM SARKAR

Abstract: Due to very high traffic in 2.4 GHz WLAN band, dual WLAN band antennas are in great demand. Printable planar antennas are the best choice in this regard due to low-profile structure as it can be best fitted in handheld devices. Antennas designed on FR4 substrate is an added advantage due to low fabrication cost. A dual band monopole antenna is suggested for potential application in wireless communication in 2.4 GHz and 5.8 GHz ISM band. It is a low profile planar antenna fabricated on FR4 substrate of dielectric constant 4.4 and width of 1.6mm. It consists of a rectangular ground plane and a semicircular patch printed on opposite sides of the substrate. The antenna resonates at 2.45 GHz and 5.8 GHz with the bandwidth of 300 MHz and 450 MHz respectively. The antenna shows positive gain and good radiation patterns at these two frequency bands. A prototype of the antenna is fabricated. Measured and simulated data are in good agreement.

Keywords: ISM Band, High Gain, WLAN, Microstrip Antenna

DETECTING INDOOR SOUND EVENTS

Toma TELEMBICI, Lacrimioara GRAMA

Abstract: In this paper we present the problem of context awareness for a service robot, based on acoustic analysis. To describe the audio signals, we proposed the liftering Mel frequency cepstral coefficients as features, while for classification the k-Nearest Neighbor is used. The results obtained are illustrated for different number of features, various filtering methods prior classification, different metrics, voting procedures and weighting methods, respectively. The best results are obtained using 37 features, City and Simple Value Difference metric, Inverse Distance voting, Accuracy Based weighting method, and k=3. The correct classification rate is improved from 98.25% to 99.21%, by applying resampling to data before classification

Keywords: MFCC; kNN; service robot; audio classification

RADIAL STUBS DESIGN METHOD FOR MICROSTRIP FILTERS BASED ON NONLINEAR EQUATION SYSTEM

Mihai Alexandru ILIE, Liviu VIMAN, Dan PITICĂ

Abstract: The theoretical background behind this paper is focused on radial stubs design for microstrip filters. A radial stub may be seen as a planar component characterized by its radius and angle. Depending on its geometrical parameters, a radial stub may have an inductive or capacitive behavior depending on frequency. Thus, a radial stub is equivalent with a series inductor-capacitor circuit. The actual paper proposes a method of radial stub design using as input parameters the equivalent capacity, inductance and the dielectric material properties. Equivalent inductance and capacity are expressed by a non-linear equation system that requires a numerical solving method.

Keywords: Radial stub, impedance, non-linear system, numeric method, microstrip, dielectric.

ANALYSIS AND COMPARISON OF UNIVERSAL gm-C BIQUAD STRUCTURES

Tomina Fabiola SALAJAN, Iulian CAMPANU, Raul ONEŢ, Marius NEAG, Marina ŢOPA

Abstract: This paper presents a comparative analysis of four multiple-input single-output (MISO) universal gm-C biquads - functional blocks able to implement second-order transfer functions such as low-pass, high-pass, band-pass, band-stop and all-pass without changes to their topology. The analysis focuses on the effects of the parasitic capacitances associated with the unloaded nodes of the biquads - nodes where no capacitance is placed - and the performance of the resulting filters. Performance comparison is based on both analytical analysis and simulation results and considers the orthogonality of the biquads, area and current consumption, how the parasitic capacitances affect the main poles of the transfer function and the capability of a differential implementation.

Keywords: universal biquad, gm-C, parasitic capacitance, transfer function, complex zero, imaginary zero

Number 1/2018

DESIGNING STRAIGHT COAXIAL CONNECTORS FOR FEEDER AND JUMPERS IN CELLULAR MOBILE BASE STATIONS

Mahdi BAHAGHIGHAT, Arash NAGHDEHFORUSHHA, Mohammad Reza SALEHIFAR, Mehdi MIRFATTAHI

Abstract: RF coaxial products are used in wide range of applications in cellular mobile telecommunication technology. In this paper, a coaxial telecommunication connector is designed for coupling to feeders and jumpers in BTS Mobile sites. The desired frequency range for this connector is 1.7 to 5 GHz. In terms of power transmission, especially when a high power is considered, one of the most important issues is the impedance matching; so designing and implementing a connector with an optimal VSWR is targeted in our work. To this end, we have tried to examine the effect of dimensional changes as well as material changes on this important parameter. Parametric study is used for various dielectric constants to show the VSWR behavior. Our simulation results clearly show that the proposed scheme can be used in practical applications.

Keywords: VSWR, Passive Connectors, Coaxial Connectors

A NEW ALGORITHM FOR FRACTIONAL BROWNIAN MOTION PROCESSES GENERATION

Paul GAJITZKI, Alexandru ISAR, Georgiana MAGU

Abstract: The fractional Brownian motion random processes occur widely in nature and are a source of considerable interest in many fields. Sometimes the acquisition of data of this kind is laborious and time consuming. This is the reason why the problem of the generation of the fractional Brownian motion processes is of interest. We present in this paper an algorithm for the generation of fractional Brownian processes based on wavelets’ theory and we analyze the proposed algorithm’s performance. We compare the performance of the algorithm with the performance of state of the art algorithms for the generation of fractional Brownian motion random processes.

Keywords: : fractional Brownian motion random process, wavelets, stationarity, communication traffic

CIRCULAR MICROSTRIP PATCH ARRAY WITH SMALL CIRCULAR SLOTS LEADING TO MULTIBAND OPERATION FOR WLAN APPLICATIONS

Rajat G. PANDEY

Abstract: This paper presents the novel design of 4 x 2 circular microstrip patch array antenna centered at 2.4 GHz. Small circular slots are subtracted from the circular patch in order to obtain the multiband operation. The goal of this paper was to obtain multiband operation, reasonable gain and SWR performance for WLAN applications. The proposed design resonates at 2.4 GHz, 3.4 GHz, 3.8 GHz, 5.45 GHz and 6.05 GHz. The size of the antenna is 24 cm x 12 cm. The proposed design includes eight circular patches of identical dimensions with equal spacing between them. Full wave simulation software was used to design and analyze the novel design presented in this paper. The results obtained of Gain, SWR and return loss are depicted in this paper.

Keywords: Circular microstrip patch, Patch array, WLAN antenna, Antenna, Mutiband

SEMANTIC ANALYSIS AUDIT IN TRIPLE-ENTRY ACCOUNTING SYSTEMS BASED ON BLOCKCHAIN

Sînică ALBOAIE, Alexandrina RAȚA, Emil HOROMNEA, Mircea VAIDA

Abstract: : Using the blockchain technologies and keeping at least three parallel accounting records, separately located, has as result the concept of triple-entry accounting. The triple-entry accounting is not an alternative for the double-entry accounting, but rather part of a solid system when the two types of accounting are combined. Decreasing the cost of managing trusty accounting by using blockchain is a significant advantage, addressing future requirements of accounting. In this paper, we are proposing an audit method based on semantic analysis of blockchain accounting data to gain resilience of accounting against improper (fraudulent) use both on the public and private sector

Keywords: blockchain, triple-entry accounting, semantic analysis, audit

Number 4/2017

LINEAR PREDICTIVE CEPSTRAL COEFFICIENTS IN WILDLIFE DETECTION SYSTEMS

Elena Roxana BUHUS, Lacrimioara GRAMA, Corneliu RUSU

Abstract: The paper presents the impact of cepstrum computation in Linear Predictive Coding for audio signals that belong to wildlife areas. In this sense, we compare on diverse number of features both Linear Predictive Coding coefficients and Linear Predictive Cepstral coefficients for the audio signal characterization. The scope is to identify the appropriate number of coefficients for both methods, together with the best method that can provide a certain accuracy within a defined framework, such as an intruder detection system. The experimental results prove that the Linear Predictive Cepstral coefficients method provides a better accuracy for tested scenarios than Linear Predictive Coding. The highest accuracy is obtained using the order of 16 for the predictor filter together with the KStar classifier.

Keywords: LPC, LPCC, KStar

AUTONOMOUS SCALE MODEL CAR WITH ULTRASONIC SENSORS AND ARDUINO BOARD

Cosmin CIUCIU, Daniel BĂRBUŢĂ, Sebastian SĂSĂUJAN, Emilia ŞIPOŞ

Abstract: An autonomous car is a vehicle that is capable to perceive the environment and navigate without human input. The paper presents the construction of an autonomous scale model car, using ultrasonic sensors and Arduino board. The designed scale model car can follow a predefined route with obstacles and turns, with the automatic change of the wheels steering angle. Data collected from the route is displayed in the smartphone application using a Bluetooth connection. The car passes the route correctly and reaches the finish line in about 30 seconds.

Keywords: Arduino board, automatically angle change, autonomous car

SECOND ORDER POLYNOMIAL FUNCTION IMPLEMENTATION USING ELIN ANALOG BUILDING BLOCKS

Mihaela CIRLUGEA

Abstract: This paper presents an ELIN implementation of a second order polynomial function. The basic building blocks of the proposed circuit consist of a multiplier, an adder and two weighting cells. These blocks were implemented using current-mode log-domain cells. The circuit was implemented using bipolar transistors and the design was validated by PSPICE simulations.

Keywords: current-mode, ELIN, arithmetic circuits

MACHINE-LEARNING BASED APPLICATION FOR STAFF RECRUITING

Anca APATEAN, Evelyn SZAKACS, Magnolia TILCA

Abstract: : The paper presents an application to classify candidates using an average of the results obtained from the use of several independent classifiers, like KNN, LDA, Tree J48 and Naïve Bayes. The main purpose of the classification is to help a company to select a large quantity of CVs by passing them through the application filter. The dataset was constructed with 172 instances (belonging to 17 classes), each characterized by 18 attributes. Information like Education, Experience, Foreign Languages, Programming Languages, Salary Range, Gender, Age and Personal Skills were considered for each candidate

Keywords: classifier, KNN, LDA, Tree, Naïve Bayes, voting, matching score

A Humidity Sensor Based on Soil Thermal Conductivity

Teodor SUMALAN

Abstract: Measuring soil moisture is one of the most important activities in multiple domains of agricultural activities and industries. This procedure helps farmers to improve their automation systems for watering more efficiently. In the other sense of meaning, soil moisture sensors are devices providing the amount of water retrieved in a volume of soil. This paper presents a prototype of moisture sensor that can measure the quantity of water in greenhouses but useful in different fields of agriculture. The sensor is able to measure the humidity of soil and transmit it to a central processing unit for further reactions. The sensor is based on dsPic30f3012 controller system and can be controlled remotely via a RS485 interface, allowing complex configurations of data acquisition and control. Despite of the imperfections of soil, containing a lot of impurities, using this sensor and the precision improvement techniques implemented, the soil humidity can be successfully evaluated.

Keywords: sensor, soil moisture, humidity, greenhouse

APPLICATION FITNESS ANALYSIS BASED ON METAMODELING TECHNIQUES – A CASE STUDY: E-BIKE

Alexandra IOSUB , Ingrid KOVACS , Andi BUZO , Jerome KIRSCHER , Georg PELZ , Liviu GORAS , Arcadie CRACAN

Abstract: Performance and behavior verification of electronic components and systems in order to fit applications requirements are indispensable nowadays. There is a particular interest in finding the application yield caused by the process variation of the electronic components and revealing the component’s factors which cause failure in the application. In order to address these two problems, the Failure Probability (FP) estimation of an electronic system is computed, and the impact of component’s factors is analyzed, by considering factors variations as well as various operational conditions. Since FP estimation can be computationally expensive, it is necessary to find more efficient methods. In this paper, we propose a method that can both estimate FP and quantify the impact of factors based on behavioral system modeling and system’s performances metamodeling. FP is calculated by re-sampling the metamodel, hence, converting a small numbers problem into a large numbers problem. A comparison between this method with other existing methods such as counting and distribution fitting is made for an E-Bike case. The results show that the approach based on metamodeling can lead to comparable results with state-of-the-art for FP estimation methods with the advantage of revealing the factors which cause the application failure.

Keywords: Application Fitness, FP, DoE, Distribution Fitting, Metamodeling, E-Bike, SystemC-AMS.

Number 3/2017

NUMERICAL MODELING OF CASCADED PHOTOVOLTAIC CELLS IN OPEN SOURCE SIMULATION ENVIRONMENT

Sergiu-Alex RANGA, Botond Sandor KIREI, Marina Dana TOPA

Abstract: This paper aims to provide and easy to understand guide to PV modeling in Scillab/Xcos. It also highlights the importance of the Newton-Raphson method and its role in plotting the P-V and I-V characteristics. The current literature is abundant in Simulink numerical models, but they lack the detailed implementation of the Newton-Raphson method. An important contribution of this paper is the graphical implementation of the Newton-Rapshon method, tailored for solving the current equation of the (cascaded) PV cells.

Keywords: Photovoltaic Array, Scilab/Xcos, P-V and I-V curve, Simulation

DESIGN FOR SENSOR BASED ON SUSPENDED CORE MICROSTRUCTURED OPTICAL FIBER

Alexandru SCROB, Jean-Louis AUGUSTE, Ramona GALATUS, Lorant SZOLGA, Nicoleta TOSA

Abstract: In the paper is proposed the design for a sensor based on a new three-hole microstructured optical fiber. The simulations show that the size of the core has a great effect on its applicability on both sensing and transmission. The core of the fiber should have the diameter around 1μm to 6 μm and the experimental results prove that the fabricated optical fiber would make a good sensor with losses measured around 4,5 dB/m.

Keywords: fiber sensor, microstructured optical fiber (MOF), cutback technique for attenuation measurements

INVESTIGATION STUDY OF FEASIBLE PRIME NUMBER TESTING ALGORITHMS

Mohammed M. ASAD, Ibrahim Marouf, Qasem Abu AL-HAIJA

Abstract: In this expository paper, we describe three primality-testing algorithms: Miller-Rabin, Fermat and AKS primality tests. The first test is very efficient, but is only capable of proving that a given number is either composite or ‘very probably’ prime. The second test is also probabilistic with lower probability and higher execution time. The third test is a deterministic unconditional polynomial time algorithm to prove that a given number is either prime or composite; however, it had no practical applications due to the time complexity O (log5 (n)). Thus, the first primality test is at present one of the most widely used in practice as it run at logarithmic run time complexity O (log (n)).

Keywords: Prime Number, Fermat Primality Test, Miller-Rabin Primality Test, AKS Primality Test

FOURFOLD MICROPHONES AREA FOR ACOUSTIC SOURCE LOCALIZATION

Olimpiu POP, Lacrimioara GRAMA, Corneliu RUSU

Abstract: In this paper we present a fourfold microphones area for source localization. The area consists of four shotgun microphones arranged at an angle of 900 between them. The distance between the microphone’s capsules can be changed in a range of 10 ÷ 20 cm. This is needed to study the phase differences between the acoustic signals captured for a given bandwidth. Experimental results based on spectrum and autocorrelation measurements have shown that this area can be used in domestic applications or for wildlife areas intruder detection.

Keywords: microphone, intruder detection, spectrum, correlation

DISCRIMINATE ANIMAL SPECIES USING CEPSTRAL COEFFICIENTS AND TESPAR ANALYSIS

Petre G. POP

Abstract: Identification of animal species based on the sounds emitted has already proven to be useful in biodiversity assessment but can also be useful in other biological research. Most researchers have used the cepstral coefficients for animal species discrimination. In this paper we explore the use of a combined cepstral - TESPAR analysis, which does not use directly the acoustic signal from animals, but cepstral coefficients derived from it (MFCC and Teager-MFCC) subject to TESPAR analysis, to discriminate between different animal species. Our experiments using this approach together with classification techniques shows that TESPAR S-matrices of cepstral coefficients can be successfully used to discriminate between different animal species, even in the conditions of small sets of training data with small length and different sampling frequencies.

Keywords: animal species identification, cepstral coefficients, TESPAR analysis, Teager energy operator, Random Forest

Number 2/2017

ENHANCED INTERACTIVITY IN VIRTUAL TOURS

Catalina MALINCHI, Radu LACATUSU, Aurelia CIUPE, Serban MEZA

Abstract: In the context of immersive environments, interactivity has become a strategy of object manipulation in customized virtual scenes. A new emerging trend has been orientated on the development of integrated solutions, where augmented reality (AR) elements are used for real space representations, specific to the estate industry and intelligent design. The current work presents a gesture-based AR application which uses physical commands to identify and manipulate 3D objects in a 360º virtual tour. A low-cost technical setup has been used for in-lab validation, as an alternative to current VR/AR headsets.

Keywords: Google Cardboard; virtual tour; explorable panorama; augmented reality; explorable panorama

DEVELOPMENT OF A SPEED CONTROL SYSTEM FOR A DC MOTOR USING A FUZZY LOGIC CONTROLER

Gabriel OLTEAN, Laura-Nicoleta IVANCIU, Horea Alin BALEA

Abstract: The aim of this work is to implement an embedded system for the speed control of a DC motor, using a microcontroller-based development board. The control system is developed around an Arduino Mega board and involves a fuzzy logic system as controller. The implementation offers an easy understanding of the main concepts regarding closed loop systems and embedded systems, operating in real environment. The main achievements of this implementation can be summarized as follows: access for data in relevant points of the system, illustration of the use of some fundamental concepts in modern electronics (magnetic hall effect sensor for RPM measurement, C++ programming, implementation of a fuzzy logic controller), possibility to explore different operating scenarios. The experimental results prove the expected operation of our speed control system.

Keywords: fuzzy system, speed control, Arduino, embedded system

PRODUCT DIFFERENTIATION AND USER ENGAGEMENT WITH AR BATTLESHIP GAME BASED APP

Dănuț VÂSC, Șerban Nicolae MEZA, Aurelia CIUPE, Aurel VLAICU

Abstract: This paper reviews the concept of augmented reality (AR) and proposes an innovative use of it in digital marketing. The purpose is to present the steps required for developing an AR app using state of the art resources and then validate its usage as a tool for product differentiation and user engagement. Approaches to user interaction and engagement stemming from the gamification area are also considered and introduced in the app design and implementation. Results regarding the proposed strategy and user feedback on the implemented app are discussed.

Keywords: Augmented Reality, digital marketing, gamification

POWER CONSUMPTION AND NOISE MARGIN COMPARISON BETWEEN SIMPLE TERNARY INVERTER AND BINARY INVERTER

Emilia SIPOS, Robert GROZA, Laura N. IVANCIU

Abstract: The paper presents a comparative study between ternary and binary inverters from the point of view of power consumption and noise margin. Both circuits are supplied with 0.9 Vdc and use the same transistors W/L ratios. Simulations show that the power consumption is smaller for ternary inverter and doesn’t reveal the output voltage edge type. The noise margin is 150mV for the ternary inverter. The results prove that the use of ternary logic is a promising method to design power attack resistant integrated circuits

Keywords: digital circuits, short-circuit power, noise margin.

Number 1/2017

FDTD MICROWAVES SIMULATOR USING ADVANCE COMPUTER GRAPHIC TECHNOLOGIES

Nicolae CRISAN

Abstract: The paper presents several technical issues for finding and rendering solutions for Maxwell’s differential equations. The finite difference time domain (FDTD) method used in the paper is the most wanted numerical analysis for microwaves rendering in the time domain. Nowadays computational electrodynamics (electromagnetics) comes with more advanced and agile techniques used in computer games. Frameworks as DirectX or OpenGL accelerate hardware using graphical processor units (GPUs) and enabling dynamic rendering in real time. These frameworks could be used to speed up the 3D rendering according to the FDTD demands in terms of the framerate, and, bring some new capabilities as rotation, translation, and scaling from computer graphic to computational electrodynamics. The presented software is intended to replace some obsolete graphic technologies or software (like Mefisto-2D or the graphic simulations based on Matlab) which are in use today in universities, for understanding the main properties of the microwaves with new and agiler rendering techniques that are coming from computer graphics world.

Keywords: Computer-aided, FDTD, DirectX, OpenGL, Maxwell’s equations, electromagnetic modeling

STATE OF THE ART: METHODS FOR VIDEO BASED FIRE DETECTION VIA STATIONARY CAMERA

Elena Roxana BUHUS, Anca APĂTEAN

Abstract: Fire outbreak is a common issue happening in open spaces (like forest areas) and the damage caused by these type of incidents is significant toward nature and human interest in matter of economy. This paper provides an analysis and comparison between several implementation methods from the literature meant to detect fire pixels regions from video input. The fundamental goal of this research is to develop an improved algorithm for fire pixel classification in video sequences.

Keywords: fire detection, RGB color model, YCbCr color model, HSV, Lab, k-means, image processing

CROWD EVACUATION USING MULTI-OBJECTIVE OPTIMIZATION AND TAKAGI-SUGENO FUZZY LOGIC SYSTEM

Laura-Nicoleta IVANCIU, Gabriel OLTEAN

Abstract: The increased frequency of events with massive numbers of attendants leads to crowd management issues. This paper proposes a system which prevents the occurrence of congestions on the evacuation routes, assuming that the number of persons per location, the sizes of the evacuation routes, and the density on each route are known. A Takagi-Sugeno fuzzy system computes the priorities for each evacuation route, based on its length and width and on the number of persons that are already on it. The allocation of the groups of people on the evacuation routes with the highest priorities is achieved through multi-objective optimization with genetic algorithms, in order to obtain a balanced evacuation of all locations, as quickly as possible. The dynamic of the number of people left in each location is displayed graphically and highlights the correct achievement of the proposed objectives.

Keywords: crowd management, multi-objective optimization, fuzzy logic

A METHOD AND SPECIFIC CHALLENGES FOR OBTAINING A HDR HYPERLAPSE FROM STILL IMAGES

Vlad CODÎRLĂ, Aurelia CIUPE, Bogdan ORZA, Șerban MEZA

Abstract: Creating an in-motion time-lapse while moving the camera over significant distances (herein named as a hyperlapse) through an environment with high variance in illuminance poses multiple challenges. Normally, one would have to choose between changing the camera settings between frames, thus causing distorted artifacts, or dealing with the compromise of selection between under-exposed or over-exposed areas during movement, that results in lower fidelity of the captured scene. The current work proposes an approach according to which each frame is batch-captured using varying exposure levels. These exposures are then combined for each frame of the hyperlapse to create a HDR (high dynamic range) hyperlapse suited for display on LDR (low dynamic range) equipment, thus combining two traditionally separate areas of research. Multiple algorithms for combining the images have been experimentally tested and compared based on several metrics: blending workflow, computational requirements and empirical quality assessment.

Keywords: HDR, high dynamic range, hyper lapse, time-lapse, experimental acquisition

ACTIVE MEASUREMENT OF THE LATENCY IN CLOUD-BASED NETWORKS

Adrian TAUT Iustin A. IVANCIU Eduard LUCHIAN Virgil DOBROTA

Abstract: This paper presents a software tool for the active measurement of the one-way delay (latency) based on cyclic-path delay method in cloud. The principle consists in sending packets from the source node to every node belonging to a multicast group. Copies of the original packet will be forwarded through every possible path in the network before returning to the source node. The original idea is represented by the breaking of the well-known flooding rules which do not allow sending back the packet to the issuing node. The software tool is composed of two modules: one in C for measuring the latency on cyclic paths (packet generation and timestamping) and one in MATLAB for the estimation of one-way delays. The equation system derived from the cyclic-path delays is underdetermined, therefore the estimation of some latencies is needed. The estimation problem is formulated as a constrained optimization problem. The calibration was performed by means of a software tool, previously developed within Unified Communications Labs (UC Labs), where all the authors are currently working. The values of the latencies have a precision of nanoseconds and a good tracking ability of unexpected delay variations. The tool is scalable for both clouds orchestrated by OpenStack and physical networks, regardless of the topology and number of nodes.

Keywords: active measurements; cloud; cyclic-path delay; latency

Number 4/2016

EFFICIENT FPGA IMPLEMENTATION OF AN ADAPTIVE FILTER FOR NOISE CANCELLATION

Ervin SZOPOS, Ioana SARACUT, Horia HEDESIU

Abstract: This paper presents an efficient architecture of the Least Mean Square (LMS) adaptive algorithm implemented as a FIR filter on a reconfigurable platform. The architecture of the adaptive filter was developed with a general structure so it can be used in many applications with minimum limitations. Besides general use, the architecture has the advantage of using optimal hardware resources on the chip. The effectiveness of the proposed architecture is demonstrated on the noise cancelling application, and the performances are evaluated in terms of signal-to-noise ratio (SNR), convergence speed, mean square error (MSE) and hardware usage.

Keywords: FPGA, LMS-FIR, architecture, convergence, SNR, MSE

EFFICIENT CHAOS-BASED CRYPTOSYSTEM FOR REAL TIME APPLICATIONS

Jean De Dieu NKAPKOP, Joseph Yves EFFA, Monica BORDA, Anton CIPRIAN, Laurent BITJOKA, Alidou MOHAMADOU

Abstract: Nowadays depending on the application domain, the need to protect images against unauthorized users has become a challenge. We previously developed and implemented using Matlab software, a new image encryption algorithm that uses the permutation-diffusion structure. The permutation step was used to shuffle the positions of the pixels within the image, while the diffusion process was employed to improve the security by masking all the pixel values of the image. In this paper we present an implementation of this chaos-based algorithm under Qt environment using the C/C++ programming language to show the practical aspects of its implementation. This implementation also provides the opportunity of selecting three encryption stages to encrypt and decrypt any type of input images. Thus, this allows us to see that the permutation-diffusion structure is required for a secure encryption. In order to prove the efficiency of this algorithm a comparison of the encryption time has been made with the other similar algorithms implemented on Matlab or C/C++. The results show that the proposed approach performs faster with higher security level for the encryption task.

Keywords: Chaos-based algorithm, image encryption; Graphical User Interface; C/C++ programming language

AN EXPERIMENTAL EVALUATION OF ANALOG TV CABLE SERVICES DISTRIBUTED IN GPON ARCHITECTURE

Radu ARSINTE, Eugen LUPU

Abstract: This paper explores the benefits of GPON (Gigabit Passive Optical Networks) in analog TV services. Analog TV service is still present in the standard triple play distribution architectures, as an effect of unique advantages: simple distribution in an apartment via standard RF splitters, unlimited number of viewing sites, real time behavior by lack of encoding/decoding processes. Of course, the quality is still limited by analogic standards, but the price/performance ratio is unbeatable. The most important parameters characterizing the analog TV performance are described, the network architecture is emphasized. To have a reference, the test architecture is a sub-network of a commercial telecom company. In this case only the quality of TV reception is performed. Using open-source software, and the integrated TV tuner board makes possible to test the levels and S/N ratio for all the analog TV channels of the spectrum. The system opens the possibility to do the same for digital channels in DVB-C standard, with minimal changes.

Keywords: GPON, analog TV, Signal to ratio, cable TV

INFRARED MONITORING SYSTEM USING MIWI TECHNOLOGY

Gheorghe TURCAN, Nicolai DOBROSTOMAT, Marius NEAG, Marina TOPA

Abstract: This paper presents a wireless data acquisition system for monitoring thermal activity using single- or multiple-point infrared sensors. The wireless capability of the system is provided by Microchip’s MIWI Point to Point protocol implemented by a PIC16LF1937 microcontroller which comprises the MRF24J40MA radio transceiver module from the same company. The received data are sent through the USB port to a PC, where they are processed and plotted by an application developed for this purpose. The application also allows the user to control all components of the systems through a simple GUI.

Keywords: IR Sensors, Thermal monitors, Wireless, MIWI, Data Acquisition, Microcontrollers

COMPARATIVE ANALYSIS OF PH CONTROL METHODS

Liliana TOMA

Abstract: The study of pH control is important for the environment as the effluent or wastewater has to be neutralized before being discharged. pH control is considered a challenge due to the highly nonlinear characteristics, time-varying nature and variations in the titration curve. This paper presents the main methods used to control the pH process. Also, it aims to provide a survey for pH control methods. This is achieved by a thorough comparison which highlights the main advantages and drawbacks for the control methods.

Keywords: adaptive control, fuzzy logic, pH control, PID, process control, neural network

EXPERIMENTING FEATURES SELECTION METHODS IN WEKA FOR COMPUTER VISION

Anca APĂTEAN

Abstract: The paper presents the main possibilities to select features by the use of Search and Ranker methods in Weka, from monomodal or bimodal vectors comprising features extracted by 8 different algorithms. How to proceed to run and interpret the results within a number of 12 FS methods, by direct or n-folds CV application. By considering different threshold values for the feature rank, a number of 48 variants of FS methods resulted and these were applied to 6 different input vectors. All these were compared by the number of selected features and by the balanced accuracy as obtained with a KNN classifier.

Keywords: features selection, crossvalidation, KNN classifier, Weka, Ranker, Search, balanced accuracy

ABOUT CONVERTING ACTIVE ANALOG FILTERS TO DIGITAL FILTERS

Lăcrimioara GRAMA, Alexandru LODIN, Corneliu RUSU

Abstract: The goal of this paper is to present the state-space approach used to obtain the digital filter directly from the netlist of the active analog filter. To reach this goal, the state-space approach has been used to obtain the state-space description of the active analog filter. Then the bilinear transformation was recalled to convert from analog domain to the digital domain. The entire approach is available software and the experimental results shown in this work have been obtained using this computer application.

Keywords: state-space, analog to digital conversion, dependent sources

AN INTEL GALILEO PLATFORM APPLICATION DEVELOPMENT USING MATLAB

Eugen LUPU, Radu ARSINTE, Horea CHIS

Abstract: MATLAB is a powerful tool often employed by researchers in order to simulate or test their applications. Most often the applications don’t interact with other external hardware, but MATLAB has facilities to communicate with external devices by the means of the I/O ports. The paper presents an approach for control applications which run on MATLAB and monitor a process interfaced to a Galileo platform. The resources provided by MATLAB allow the user to change and experiment very easily with different control algorithms and to benefit of a very friendly GUI for the applications. This type of application is targeted to educational purposes.

Keywords: MATLAB, educational application, Intel Galileo platform, control algorithm, Arduino

QUANTIFICATION AND PREVENTION TECHNIQUES OF PROSTHESIS-PATIENT MISMATCH

Radu A. SASCĂU, Cristina OLARIU, Cristian STĂTESCU

Abstract: Aortic valve surgery is now the second most commonly performed cardiac operation because nowdays is found an increasingly elderly population. Echocardiographyc examination (in particular Doppler assessment) is the standard clinical technique for quantifying of hemodynamics after surgical correction of valvular lesion. Analysis can include specifics parameters: mean and peak transprosthetic gradients, dimensionless valve index and effective orifice area (EOA) from the continuity equation. Most patients indication for aortic valve surgery will ‘technically’ have some degree of patient – prosthesis mismatch (VP- PM), but this is sometimes severe. Understanding the techniques by which
echocardiographic imaging and Doppler technique are used to noninvasively assess aortic prosthesis hemodynamics, and the caveats associated with those methods, are useful for cardiologist to differentiate causes of high tranprosthetic gradients and prosthesis dysfunction from other causes of obstruction. EOA indexated should be measured in the first 3 weeks postoperatively or at hospital discharge. Assessing the presence and severity of the phenomenon of mismatch is important because that will affect long-term outcomes. Use of the EOA indexated as a continuous variable may help to define the level of severe VP–PM that results in increased mortality, and this may occur at a critical level of obstruction. Prediction of severity of VP–PM is difficult to estimate, that why, the primary goal should be to prevent severe VP–PM by the use of new generation aortic prosthesis with superior haemodynamics and approach with modern surgical techniques. This article summarizes the means by which echocardiography/ Doppler technique is used to assess hemodynamics after aortic valve replacement and the potential caveats associated with its use, and proposes an algorithm for the clinical evaluation of high gradients after aortic surgery.

Keywords: : echocardiography technique, aortic valve replacement, mismatch phenomenon, effective orifice area, transprosthetic gradient

Number 3/2016

DESIGNING EFFICIENT MULTIMODAL CLASSIFICATION SYSTEMS BASED ON FEATURES AND SVM KERNELS SELECTION

Anca APATEAN

Abstract: An efficient classification system uses only the most representative features extracted from images in order to reach a decision. A multimodal system may consider multiple sources of such information. Selecting those features is not a simple task due to the fact that multiple features selection (FS) methods exists, with multiple setup possibilities and multiple possible feature vectors to be applied on. Moreover, by applying the FS, the new vector may comprise too few features and the recognition accuracy to significantly drop. This paper proposes solutions to compensate that accuracy loss by the SVM kernel selection.

Keywords: object classification system, features selection, features extraction, SVM kernel selection, multimodal

AUTOMATIC PARKING ACCESS USING OPENALPR ON RASPBERRY PI3

Elena Roxana BUHUS, Daniel TIMIS, Anca APATEAN

Abstract: Automatic parking access using a real time embedded system for Automatic License Plate Recognition (ALPR) and access control is common these days in urban commercial spaces, office buildings and other similar public spaces. This paper provides another method for implementing an ALPR system using an open source C/C++ library called Open ALPR, based on OpenCV and Tesseract- OCR. The fundamental goal of the system is to use image processing to identify every single vehicle that goes in and out through the barrier gate parking system. The hardware is implemented with the latest version of Raspberry Pi, and a Pi Noir camera.

Keywords: OCR, Vehicle license plate, OpenALPR, Application Programming Interface (API), Raspberry Pi

COMPARISON OF SENSITIVITY ANALYSIS METHODS IN HIGHDIMENSIONAL

VERIFICATION SPACES

Ingrid KOVACS, Marina TOPA, Andi BUZO, Monica RAFAILA, Georg PELZ

Abstract: The modern electronic systems grow in complexity as the number of potential factors that affect the systems behavior is increasing. The analysis of such systems can be simplified if the most important factors are identified upfront. Sensitivity analysis addresses this problem. The paper describes and compares six sensitivity analysis methods based on: variance decomposition, One-Factor-at–a-Time screening and regression. The methods are applied on a set of custom functions and a real system from the field of automotive electrical systems: a battery control model. The best performances were obtained by the following methods: the variance-based methods (Jansen and EFAST) and the regression based method.

Keywords: Sensitivity analysis, design verification, factor prioritization, variance decomposition, regression

OUTSPREADING ENTERPISE CAPABILITIES INTO THE CLOUD – A COMMERCIAL CASE STUDY

Sorin POPA, Mircea-Florin VAIDA

Abstract: Gravitating around the ERP integration concept, the present paper aims to introduce a case study of a world-wide distributed commercial solution, expanding into the cloud to provide ERP empower enterprises with mobility, process automations or workflow management. Motivated by ERP vendor independence and global commercial adoption, its objective is to provide a generic integration strategy, easily scaling between back office environments. Expanding on history and evolution, the paper presents some of the most important integration options, concluding on the adoption results and the strategy perspectives.

Keywords: ERP, cloud, integration, process automation, connectivity

FUZZY MODELLING OF THE VOLTAGE TRANSFER CHARACTERISTIC FOR THE SIMPLE TERNARY INVERTER

Emilia SIPOS, Laura IVANCIU, Lavinia MICLE

Abstract: Given the fact that the design and especially the manufacturing of some ternary circuits are now possible, using the basic ternary circuit to design more complex ones becomes a necessity. The simulation using models lead to savings of time and money in the design process. The aim of the paper is to model, using fuzzy logic systems, the voltage transfer characteristic for simple ternary inverter, the basic circuit in the design of complex ternary circuits. The obtained fuzzy system provides a correct voltage transfer characteristic.

Keywords: Fuzzy logic system, ternary inverter, voltage transfer characteristic

OPTICAL SYSTEM USED FOR SORTING APPLES BY SIZE

Lorant SZOLGA, Alexandru SAMSODAN

Abstract: A low cost sorting system by size for apples is presented. This kind of system will be very useful in a more complex sorting machine, where is appropriate to sort the products on their color, dimension or weight. The development of the system relies on the NIR (near infrared) technology and linked with microcontroller data processing. We proposed this technology as an alternative to the more expensive video processing systems. The infrared components were distributed in a 3D fashion in a chamber with black walls. The communication results between the sensors were interpreted by our proposed algorithm for the volume calculus of the apples. The system was tested on different sets of apples and proved to be a reliable one in each case. The final prototype can be used not only for sorting but for calibration too, thus permitting a more accurate sorting. This work will be a good starting point for the next building block of the sorting machine that we are working on

Keywords: NIR (near infrared), sorting, calibration, microcontroller, LED, photodiodes

Number 2/2016

DESIGNING AND IMPLEMENTATION OF A PV POLYCRYSTALLINE PANEL-BASED, TWO-AXIS SOLAR TRACKING SYSTEM

Anca APĂTEAN, David ABRUDEAN

Abstract: A Solar Tracking System (STS) is developed to minimize the angle of incidence between the sunlight and a photovoltaic (PV) panel surface. The paper presents the designing and implementation aspects in order to obtain a system capable to automatically direct a solar panel to the point of maximum light intensity, considering two axes: North-South (for differences in angle during winter vs. summer) and East-West (over a day). By using the tracking system we developed, the total power generated by the panel is higher than the one obtained by a fixed configuration, the panel efficiency being estimated to increase with 42%.

Keywords: photo-voltaic polycrystalline panel, solar energy, two-axis tracker, photodiode, motors, PCB design, Arduino

AN INTELLIGENT EYE-DETECTION BASED, VOCAL E-BOOK READER FOR THE INTERNET OF THINGS

Anca APĂTEAN, Flaviu B. DUNCA

Abstract: This paper presents an intelligent vocal e-book reader based on the Raspberry PI, that is remotely controllable through a web interface, being aimed for the Internet of Things. The software was mostly written in Python, but knowledge about HTML/CSS was also used. Images of the user are processed for face and eye detection by classification using LBP image features within OpenCV. To allow the system to accurately function under low light conditions, a light sensor is used in conjunction with a series of leds which are lit up when the lighting in the surrounding room is poor.

Keywords: Internet of Things, machine learning, classifier, intelligent application, image features extraction, eye detection algorithm, Raspberry PI, web interface, Python programming

AN ADAPTIVE VISUAL DICTIONARY BASED APPROACH TO OLD FILM RESTORATION

Mihaela CISLARIU ,  Mihaela GORDAN , Aurel VLAICU , Monica BORDA

Abstract: One of the major problems for the old films is their degradation due to aging. Over recent years there has been an encouragement for the development of virtual restoration tools, many of them using image processing. In this paper we present a new method of defect detection and virtual restoration of digitized old films frames. The proposed algorithm includes: a machine learning based small defect detection, combined with an adaptive matching pursuit algorithm for the small defects removal. The results are presented showing the good performance of the proposed solution and its suitability for virtual restoration of some digitized films

Keywords: old movie, matching pursuit, fuzzy logic, logistic model trees, defect detection, defect correction

SLIDING MODE CONTROL OF BUCK CONVERTERS FOR IMPROVED TRANSIENT RESPONSE

Cristian ORIAN, Dorin PETREUŞ, Radu ETZ

Abstract: The following paper describes a method to design a sliding mode (SM) controller for a dc-dc converter. It focuses on producing an aperiodic response and a controlled settling time. The method can be used for a given set of disturbances, and thus lends itself to subsequent worst case or weighted convex optimization solutions. The impact of an analog to digital converter is presented. Additionally, the restriction of the switching frequency, a challenge particular to sliding mode control (SMC), is discussed. The work comprises Mathcad modeling, Psim simulation and experimental results. The control block was implemented on an Artix 7 FPGA

Keywords: sliding mode control, aperiodic response, time costant, dc-dc buck converter, nonlinear control

EXPERT SYSTEM FOR RECEIVER ARCHITECTURE SELECTION

Laura-Nicoleta IVANCIU, Gabriel OLTEAN

Abstract: Selecting the architecture of a radiofrequency receiver is most often a truly tedious task, as it involves certain constraints that need to be met. This paper proposes an innovative method for choosing among various types of architectures, using a fuzzy expert system. It works with a set of rules, derived from the known characteristics of a finite number of possible receiver structures. The system suggests the architecture that best fits the requests, and also justifies the answer. The proposed solution makes a perfect tool for beginners or for educational purposes, as it also provides explanations regarding the reasoning process. The system proves to be a reliable and robust one, as demonstrated through various testing scenarios.

Keywords: expert systems, fuzzy logic, RF receiver architecture

COMPARATIVE ANALYSIS OF DIFFERENT STRUCTURES OF CHAOS-BASED CRYPTOSYSTEMS: A SURVEY

Joseph Yves EFFA, Jean De Dieu NKAPKOP, Mihaela CISLARIU, Monica BORDA

Abstract: In recent years the interest for image encryption based on chaotic maps has become a major concern. In this paper, we evaluate each step of the process used in image encryption based on chaotic maps. In an attempt to give an answer to each of the issues raised, we perform experiments and we analyze performances of a few chaos-based cryptosystems by using common analytical criterions like entropy analysis, correlation coefficients, key space, key sensitivity, number of rounds, etc. By evaluating the results, the study shows that majority of schemes that rely on converting chaotic values in real numbers are time consuming. For real-time applications, it is important to improve image encryption schemes by reducing the cyclic digitization of chaotic numbers in the generation of permutation and diffusion keys.

Keywords: Image encryption, permutation only, diffusion, permutation-diffusion, combining permutation and diffusion

SECURE ACCESS SYSTEM ON THE QorIQ NXP PLATFORM USING RSA AND DSA

Raul MALUTAN, Mircea POP, Mihaela CISLARIU, Monica BORDA

Abstract: Secure data transmission can be achieved using encryption algorithms. Combining these algorithms with authentication or communication protocols one can achieve a more secure system. Using at one of the ends a configurable platform, the system will be more scalable and the advantages will increase by adding multiple options for extension. The work proposes a secure communication ensured by RSA algorithm together with DSA between a NXP QorIQ P1025 board or an x86 machine as client and an x86 machine as server. The NTP protocol was implemented to compare the clock from the client with the clock from the server.

Keywords: security, RSA, OpenSSL, authentication, encryption, decryption, semiconductors

Number 1/2016

MIGRATION OF AN SDN-BASED TESTBED INTO A PRIVATE CLOUD: AN OPENSTACK APPROACH

Eduard LUCHIAN, Iustin A. IVANCIU, Andrei B. RUS, Gabriel LAZAR, Virgil DOBROTA

Abstract: The paper discusses an example for the migration of a Software-Defined Networking-based testbed into a private cloud. The software solution is entirely open-source and starts from an existing 4-node real implementation in OpenFlow. The OpenStack approach employs the same topology, while running OpenvSwitch on each virtual node. The Beacon controller which governs the SDN implementation is also virtualized. Preliminary results show that the migration is feasible and also it is very promising for the need of composing systems of systems in the Future Internet.

Keywords: OpenFlow; OpenStack; private cloud; Software-Defined Networking

ZIPMAP: LOCALIZING MULTIPLE POINTS OF INTEREST - MAPS AND STATISTICS

Julia FERENCZ, Orsolya PENZES, Gyula USZKAI, Florina DUDU, Evelin HORVATH, Florian Mircea BOIAN, Remus ORASAN

Abstract: Geocoding has gained in recent years a great importance in fields as environment, ecology, engineering and public health. There have been software applications developed, which, using geocoding, generate different maps based on postal codes or coordinates considering users’ preferences, but rarely conduct further various analysis. Localizing points of interest on a map and analyzing then the coverage area can be the start of a wide variety of research and statistics. These rare applications can be usually downloaded for a fee. Our application, ZipMap, is a free web based application which can be accessed and used for generating different types of maps using postal codes as points of interest. Using postal codes to generate maps is more user friendly, because this information is more accessible and less time consuming then using latitude and longitude. Also, the maps can be divided in smaller units, such as Federal States, counties, districts and also postal codes. Beside map generation, individual statistics can be generated regarding different incidences. Excel files can be downloaded containing the number of hits for postal code and theoretical incidences regarding geographical areas and population. The field in which this application has been used until now is the medical field. Hospitals can upload their patients’ postal codes. The generated maps then show the coverage area of a hospital, which are the locations from where most patients come to the specific hospital. This can be analyzed and the hospital can then take actions in order to absorb more patients to the unit. This is just an example of how useful our applications’ results can be in fields that influence everyday life quality or important marketing or state decisions.

Keywords: maps; point of interest; coverage area; zip/postal codes; geocoding

A BAND PASS FILTER FOR THE 436 MHZ BAND

Adrian DONE

Abstract: To reduce interference, on the satellite receivers input, in the signal band the intercalation of a filter with the smallest possible insertion loss (less than 1 dB) is necessary , but with an attenuation out of useful band as high as possible. In designing, a series of data was obtained through simulation that allowed the practical execution of some filters. The filter series is composed of a filter with one l/4 coaxial segment, with 3 l/4 coaxial cells capacitive coupled, a 5 cell interdigital filter inductively coupled, with fixed input and output, and a capacitive coupled interdigital filter adjustable both at the input and output.

Keywords: filter 435 MHz, coaxial filter, interdigital filter

DEVELOPMENT OF THE NEW ELF/VLF RECEIVER FOR DETECTING THE SUDDEN IONOSPHERIC DISTURBANCES

Le MINH TAN , Keyvan GHANBARI

Abstract: The Extremely Low Frequency (ELF; 3 – 3000 Hz) and Very Low frequency (VLF; 3 – 30 kHz) technique is useful for studying the D-region ionosphere of the Earth. Based on our previous receiver, we developed a new ELF/VLF receiver with three main parts: loop antennas, amplifier and data logger. The antenna includes two loops which are right isosceles triangles with the base of 2.6 m. Each loop consists of 8 turns of 18 AWG copper wire. A surface of one loop orients the direction of North - South, the surface of other loop orients the direction of East - West. The developed amplifier with two channels has the flat frequency response in the frequency range of 1 – 20 kHz. The amplified signals are digitized by a standard sound card with the sampling rate of 96 kHz and calibrated by 1 PPS of the GPS receiver. UltraMSK software has been used to detect the Sudden Ionospheric Disturbances (SIDs). To demonstrate the ability of instrument, we recorded and analyzed the SID events during 2013 - 2014 at a low-latitude station, Tay Nguyen university (TNU), Vietnam (Geog. 12.65o N, 108.02o E). The results in the daytime Wait’s

parameters are in accordance with the other authors’ results.

Keywords: ELF/VLF receiver, frequency response, EIWG, tweek atmospherics, solar flares, SIDs, Wait’s parameters

CHURN PREDICTION IN THE TELECOMMUNICATIONS SECTOR USING NEURAL NETWORKS

Ionut B. BRANDUSOIU, Gavril TODEREAN

Abstract: One of the main concerns of telecommunications companies is the subscriber retention. This sector has an annual churn rate of approximately 30% and is on top of the list. Since it is more costly to acquire a new subscriber than to retain an existing one, it is important for these telecommunications operators to identify subscribers that are at risk of churning by using predictive models. In this paper, we present an advanced methodology that predicts churn in the pre-paid mobile telecommunications industry by applying our own neural network architecture on a dataset that contains call details records. We use the gain measure and the ROC curve to evaluate the model.

Keywords: call details records, churn prediction, neural networks, telecommunications

SUPERVISION OF A PHOTOVOLTAIC SYSTEM FOR MAKING IT TRANSPARENT IN UNPREDICTABLE WEATHER

Serge Maxime FONOU, Dorin PETREUS, Martin KAMTA, Khalil KASSMI

Abstract: It is well known that the photovoltaic source remains intermittent, since it depends on unpredictable weather. In order to overcome this inconvenient in the case of standalone photovoltaic system, a solar battery is often used. However, the black dot of a photovoltaic system with battery is the guarantee of their lifetime according to manufacturer's forecast. This lifetime is uncertain when the photovoltaic system is supervised by users. In this paper, a supervising algorithm of a photovoltaic power supply source has been developed and simulated in Proteus environment for making it transparent to the users.

Keywords: Microcontroller system, Standalone photovoltaic system, Numeric simulation, Management of photovoltaic energy.

Number 4/2015

ON DATA ORGANIZATION AND COLLECTION POLICIES IN THE INFORMATIONAL SYSTEMS OF THE GERMAN CANCER SOCIETY’S CERTIFICATION PROGRAM

Julia FERENCZ, Christoph KOWALSKI, Victor BADIC, Orsolya PENZES, Sebastian DIENG, Andreas KÄMMERLE, Simone WESSELMANN

Abstract: In order to improve the quality of treatment for cancer patients, the German Cancer Society implemented the certification of cancer centers, including rigorous data collection. The data are reviewed following a systematic collection-verification-storage-reporting pathway to make them usable for analyses of care quality across centers and improving of the certification criteria. The assessment of the input data from the centers leads to several output products for the centers. Our example shows that high quality data managing processes in a newly-established, growing certification system need to be constantly reviewed and formatively evaluated over time, requiring lots of efforts and resources.

Keywords: German Cancer Society; data collection; database; benchmarking; health care; certification

AN ENHANCED PRE-AMPLIFIER FOR COCHLEAR IMPLANTS

Vipan KAKKAR, Suhaib AHMED, Sakshi KOUL

Abstract: Analog pre-amplifier for microphone used in cochlear implants, having improved biasing technique has been described. This low-noise circuit relaxes the ADC design requirement, which is designed to replace the input stage of existing Σ/Δ modulators, and has a differential voltage output. Except for the microphone, it needs no external components and uses only two input pins. This input stage achieves a signal to noise ratio of better than 80dB in a frequency band of 100 Hz-10 kHz at a 90dB SPL input. The Total Harmonic Distortion of the circuit is around -60dB at a 90dB SPL input signal with a frequency of 1 kHz. The power consumption of the input stage is 74μA excluding the microphone DC bias current. This circuit is primarily designed for use with microphone cartridges for cochlear implants.

Keywords: Biasing circuit, Distortion, Electret microphone, Low Power, Sigma-Delta Modulator

THE GENERATION OF THE RESONANCE PHENOMENON IN ACCUMULATORS THROUGH CHARGING PULSES

Levente SZABÓ

Abstract: Inside the process of charging batteries with pulses, besides the desulfation phenomenon, there is also another phenomenon, that is caused by the abrupt slope of the pulse. This fact is shown in the signal forms obtained from precise voltage measurements between the battery electrodes. During the transition periods of pulses at microscopic level there is a succession of consecutive and parallel processes, that change from the state of equilibrium to state of charge. If we choose properly the pulse for charging, the system can go in state of resonance.

Keywords: Resonance With Pulses, Charging Of Accumulators, Lead-Acid Battery

IONOSPHERIC HF CHANNEL MODELING AND END-TO-END HF SYSTEM SIMULATION

Oana Aronia MURESAN, Andra PASTRAV, Emanuel PUSCHITA, Tudor PALADE

Abstract: HF frequency band, lying within 1.6-30 MHz, is very important from data communications point of view and is unique in its property that it is refracted by the ionosphere. The main advantage of HF communications is their ability to enable data transmission over very long distances using ionospheric propagation paths without any preexisting infrastructure, making them suitable for many applications. This paper presents a detailed study regarding the modeling of a low latitudes and moderate conditions channel and its integration in an end-to-end system using the software environment VSS from National Instruments in co-simulation with MATLAB. This channel model considers that the channel fading is characterized by a Rayleigh distribution, which assumes that the magnitude of a signal passing through a communications channel will vary randomly. The ionospheric channel model is simulated through the MATLAB block, using the specific parameters of the low latitudes and moderate conditions channel, according to the Recommendation ITU-R F.1487. The simulation results illustrate the received signal attenuation level at the receiver after its propagation through the modeled ionospheric HF channel.

Keywords: HF channel modeling, ionosphere, co-simulation, NI AWR VSS, MATLAB

DETERMINATION OF THE TEC PARAMETER BASED ON GPS RECEIVED SIGNALS

Alexandru POSTINIUC, Andra PASTRAV, Emanuel PUSCHITA, Radu ARSINTE, Tudor PALADE

Abstract: The scope of this paper is to provide a method to determine the variation of the TEC of the Ionosphere by means of GPS received signals. The solution makes use of dual carrier phase and pseudo-range measurements from a terrestrial receiver, stored in specific RINEX files. The implemented algorithm processes the data and provides the variation of TEC and TEC gradient. For a considered period of time, the TEC variation indicates the behavior of the Ionosphere allowing a better modeling of the radio channel.

Keywords: TEC parameter, Ionosphere, channel modeling, GPS signals, MATLAB implementation

ELIN PROGRAMMABLE GAIN AMPLIFIERS

Mihaela CIRLUGEA , Lorant SZOLGA

Abstract: This paper presents the ELIN (externally linear internally nonlinear) method for designing a PGA (programmable gain amplifier) using MOS multiplying and weighting cells. This procedure is useful, for instance, in case of VLSI support vector machine classifiers that require many circuits for performing calculations, with high density but need good performances. This method is useful for SVM architectures because it requires simple nonlinear cells for an extended operation range. The procedure is good for weighting from 2⁰ to 2^-6.

Keywords: current-mode, ELIN circuits, log-domain, PGA (Programmable Gain Amplifier), modular analog VLSI design, LIN-ELIN transformations.

CLUSTERING FOOD RECIPES USING A FLOWER POLLINATION-BASED METHOD

Viorica R. CHIFU, Cristina Bianca POP, Ion SALOMIE, Andreea ACZEL

Abstract: This paper models the clustering of documents containing food recipes as an optimization problem which is solved using a Flower Pollination-based method. The first step of this method consists in preprocessing the documents describing food recipes using natural language techniques, while the second step consists of representing the preprocessed documents using vector space models. The vector space model representations of the documents containing food recipes are further submitted to the Flower Pollination-based clustering algorithm which clusters the documents based on their content similarity. The content similarity between two documents is evaluated using the cosine similarity measure. The proposed method has been evaluated on a set of food recipes documents available in the RecipeML format.

Keywords: food recipes clustering, Flower Pollination Algorithm, tf-idf, cosine similarity.

Number 3/2015

USING A PLC TO COMMAND A PLASTIC INJECTION MOLDING MACHINE

Anca APATEAN, Carmen MISAROS

Abstract: In the present paper, a control system for a Plastic Injection Mold Machine was proposed by using a Programmable Logic Controller (PLC) of type IMO iSmart ED-RD-20. The PLC is intended to be implemented in an automated industrial system to yield plastic products, such as flowerpots. For the practical implementation of a prototype of the system, besides the PLC, a power source (24 V DC), a temperature sensor, two DC motors, LEDs for signaling the inputs and outputs states, switches and buttons were also used. The project was realized in the Ladder Diagram (LD) language, comprising 8 LD programs and was simulated with the software simulator SMT.

Keywords: programmable logic controller, ladder diagram, plastic injection molding machine

WEB HARVESTING AND SENTIMENT ANALYSIS OF CONSUMER FEEDBACK

Emil St. CHIFU , Tiberiu St. LETIA , Bogdan BUDISAN , Viorica R. CHIFU

Abstract: This paper presents a technique for detecting, extracting and evaluating customer reviews given to online products or services. The technique proposed consists of a web harvesting algorithm and a sentiment analysis method. The web harvesting algorithm extracts customer reviews from Web pages. It extracts reviews given to online products or services, while the sentiment analysis method uses the data returned by the web harvesting method in order to evaluate the opinion polarity of the customer reviews. The technique proposed is intended to work generically, for the reviews and feedback of any given list of URLs which point to web pages where such data is available.

Keywords: feedback, web harvesting, sentiment analysis, opinion-bearing, SentiWordNet.

DESIGN AND TESTING OF AN AUTOMATED RECEIVING SYSTEM FOR THE IONOSPHERIC SOUNDING IN HF RADIOFREQUENCY RANGE

Paul BECHET, Radu Mircea SCORTAR, Teodor TODOROV, Blagovesta BONEVA, Simona MICLAUS

Abstract: The paper presents a system designed for automatic measurement of signal level in ionospheric channels based on a spectrum analyzer controlled by a software application developed for fast and flexible sounding. It enables choosing of desired frequencies, bandwidths, times and duration of monitoring and customized recordings. The measured values of two ionospheric channels were compared against the ones provided by the prediction of the commercial application Voice of America Coverage Analysis Program (VOACAP). Sounding recordings proved that automatic measurement is able not only to assess realistic received power versus time continuously but also to distinguish fading periods.

Keywords: continuous ionospheric channel monitoring, signal to noise ratio, ionospheric channel power, HF automatic receiving system

AN ECG ARTIFACT REDUCTION METHODOLOGY BASED ON PARAMETER ADAPTATION OF ANALOG FILTERS

Paul FARAGÓ , Claudia FARAGÓ , Sorin HINTEA

Abstract: Portable and wearable bio-medical monitoring systems operate towards in-vivo acquisition and analysis of bio-potentials in ubiquitous healthcare. A challenge which needs to be addressed is artifact suppression. This article proposes a solution for electrocardiogram (ECG) artifact reduction via tunable analog filtering. An electronically tunable operational transconductance amplifier was developed to implement a Kerwin-Huelsman-Newcomb (KHN) biquad. A filter parameter adaptation methodology was implemented on a feed-forward control loop, to determine the filtering characteristics for frames of ECG signal. Simulation results validate the proposed methodology for ECG artifact reduction.

Keywords: ECG analog front-end, operational amplifiers, Kerwin-Huelsman-Newcomb (KHN) biquad, feed-forward control, parameter adaptation

DEVELOPMENT OF A HOME TEMPERATURE CONTROL UNIT BASED ON A MODEL PREDICTIVE CONTROL STRATEGY

Ciprian LAPUSAN, Radu BALAN

Abstract: The paper presents the development of a home temperature control unit that aimed to replace standard residential building temperature control units with one which has superior performance and improves the comfort for occupants. The new developed system integrates a model predictive control strategy and building model online identification algorithms that assure a large applicability for different buildings. The implementation of the control unit is performed using the Discovery development board. For testing the control unit a Hardware in the loop simulation experiment is developed using dSpace platform. The obtained results showed that the proposed systems have better results from energy efficiency point of view.

Keywords: temperature control unit, rapid control prototyping, MPC control strategy

SPEAKEASY-A MOBILE ASSISTANT FOR THE MANAGEMENT OF SOCIAL ANXIETY IN SPEECH SITUATIONS

Kuderna-Iulian BENTA, Marcel CREMENE, Silviu-Andrei MATU, Ioana Roxana PODINA

Abstract: In the context of the exponential growth of portable devices, mobile applications ease the access to psychological interventions anytime and everywhere. SpeakEasy is a mobile assistant for the management of social anxiety in public speech situations, a highly distressing psychological disorder and targets the most investigated mechanisms underlying corresponding to it. It includes self-evaluation and gradual exposure to feared situations via dedicated guidelines and in situ interventions in various mobile environments. The graphical interface, progress and reward user feedback mechanisms are design features implementing the SpeakEasy model. Each mobile phone database records user’ evaluations and sends reports to the core database where psychologists can analyse them in a web-based interface

Keywords: Mobile Application, Affective Computing, Pervasive Computing, Social Anxiety

SIMULINK MODEL FOR A BIO-DIESEL ELECTRICAL POWER GENERATOR USED INTO A FARM

Alin GRAMA , Dorin PETREUS , Toma PATARAU , Radu ETZ

Abstract: The paper presents a Simulink model for a bio-diesel electrical power generator used to supply a farm. Considering that for a farm is too costly to connect to the national power grid, is taken into account electricity production using a diesel generator powered by bio fuel. Most systems of electricity generation using bio-diesel engine works all the time regardless of the rated electrical load. To avoid unnecessary fuel consumption when the electrical load is small was made an adjustment of mechanical power according with the electric load. For this purpose it was used a governor. As electric generator was used a synchronous machine because it can generate voltage with a constant frequency if the rotational speed of the rotor is constant. Electrical load was modeled using electric power consumption data collected from a real farm. Finally, a Simulink/Matlab model was created, which can be useful to engineers to design a real system.

Keywords: bio-diesel, Simulink, synchronous machine, renewable energy

AVERAGED BINARY SPARSOGRAM FOR WILDLIFE INTRUDER DETECTION

Lacrimioara GRAMA, Corneliu RUSU

Abstract: This paper presents a method for wildlife intruder detection. For the feature extraction step, we have used the averaged binary sparsogram, while for the classification step we have used artificial neural networks. Two frameworks are discussed: in the first framework, four sound classes are considered (birds, chainsaws, tractors, speech), and in the second framework, five sound classes are considered (birds, chainsaws, tractors, speech, gunshots). High overall accuracy classification rates were obtained, for both considered frameworks: 98.8% for the first framework, and 97.4% for the second one. For each framework, two scenarios are tested: first the classification is performed in one step, and second the classification is performed in two steps. A small improvement in the overall accuracy classification rates is observed when the classification is performed in two steps.

Keywords: intruder detection, short time Fourier transform, sparsogram, artificial neural network, pattern recognition

IPV6 EXTENSION OF HTB-TOOLS FOR LINUX TRAFFIC CONTROL BASED ON HTB

Doru Gabriel BALAN, Alin Dan POTORAC, Adrian GRAUR

Abstract: It is well known that traffic control is a permanent challenge for network engineers. The present paper proposes a novel method for Linux QoS (Quality of Service) solution used in network management. The research is focused on the classful queueing disciplines and specific HTB (Hierarchy Token Bucket) Linux implementations. This paper proposes a new practical improvement of bandwidth management software, named HTB-tools. These tools are generally used for implementing QoS policies based on the HTB algorithm, under a common Linux environment. The proposed improvement of the HTB-tools packages updates the software components with features that allow the implementation of QoS solutions for computer networks running the latest Internet Protocol version, IPv6. Our solution is extending the applicability domain of the HTB-tools QoS instrument to the addressing space of the IPv6, deploying an IPv6-ready rule parser component.

Keywords: communication system traffic control, quality of service, queueing analysis, IP networks

Number 2/2015

CONSTRUCTING A LOW-COST ELF/VLF REMOTE SENSING TO OBSERVE TWEAK SFERICS GENERATED BY LIGHTNING DISCHARGES

Le MINH TAN, Marjan MARBOUTI and Keyvan GHANBARI

Abstract: The Extremely Low Frequency (ELF) and Very Low frequency (VLF) technique are useful for studying the lower ionosphere of the Earth. Exploring this region has been begun in Viet Nam. We constructed the ELF/VLF remote sensing with low cost. This system includes a loop antenna, a pre-amplifier, a sound card, a PC with software and a GPS receiver. After surveying its characteristics, we found that the antenna has the high sensitivity and the pre-amplifier designed has the flat frequency response with frequency from 1 kHz to 20 kHz. It can record sferics generated by lightning discharges to estimate the parameters of the nighttime D-region ionosphere. It is also a teaching module to educate students in Space physics.

Keywords: ELF/VLF remote sensing, normalized antenna sensitivity, frequency response, tweek sferics, D-region ionosphere.

ADAPTIVE COMBINATION OF SECOND ORDER VOLTERRA FILTERS WITH NLMS AND SIGN-NLMS ALGORITHMS FOR NONLINEAR ACOUSTIC ECHO CANCELLATION

Cristian CONTAN, Mircea VAIDA, Tudor PALADE, Marina TOPA

Abstract: : In this paper, starting from a robust statistics (RS) adaptive approach presented in a previous work entitled the combined NLMS-Sign (CNLMS-S) adaptive filter, an automatic combination technique with similar performances is proposed. Thus, in order to obtain better performances in acoustic echo cancellation (AEC) setups than with the normalized least-mean square (NLMS) algorithm, in the CNLMS-S case the decision between the two algorithms (NLMS and Sign) is based on a set error threshold. The error threshold can be empirically determined or known a priori if the signal-to-noise ratio (SNR) value from the loudspeaker-enclosure-microphone (LEM) setup is available or if the local noise levels can be determined from the silences. Here, to overcome this shortcoming, an adaptive combination of the two algorithms involved in RS is highlighted, providing similar results regarding convergence and final misadjustment. Also, the need of the error threshold set by the user is removed, the combination being controlled only by a step-size parameter, independent on the LEM, constrained only by the stability range. The proposed method is compared to the CNLMS-S in nonlinear LEM setups using measured linear and quadratic Volterra kernels, tracking the behavior of the echo-return loss enhancement (ERLE) characteristic. As input sequences, audio signals with different PDFs are used and WGN is added as local noise. Simulation results justify the efficiency of the proposed method, both in convergence and steady-state error against the CNLMS-S and, implicitly the NLMS and the Sign-NLMS algorithms.

Keywords: Acoustic echo cancellation, adaptive filters, Volterra models, the NLMS algorithm.

ISA100.11a CAPABILITIES FOR INTRA-SPACECRAFT COMMUNICATION

Ioan DOMUTA, Tudor PALADE, Emanuel PUSCHITA, Ovidiu RATIU, Radu ARSINTE

Abstract: The main characteristic of the communication between spacecraft subsystems is the communication reliability. The performance growth of the low power wireless network may lead to their use for avionics communication. The ISA100.11a standard by the time, frequency and space diversity allows the implementation of a robust network and can reach the communication requirements for the spacecraft equipment. This paper analyses the resource allocation in an ISA100 network, for achieving the same throughput and latency as existing wired MIL1553B or CAN buses. Based on the this study the ISA100 network, on the current specifications, does not achieve the spacecraft communication requirements, but by adopting the UWB as the physical layer it can become a strong competitor for intra-spacecraft communication.

Keywords: ISA100.11a, wireless, spacecraft communication, MIL1553B, CAN bus

LRPT WEATHER SATELLITE IMAGE ACQUISITION USING A SDR-BASED RECEPTION SYSTEM

Paul Vladut DASCAL, Paul DOLEA, Octavian CRISTEA, Tudor PALADE

Abstract: The LRPT (Low Rate Picture Transmission) direct broadcast service provides imagery from polar-orbiting weather satellites directly to end users. As a digital transmission system within VHF band, its main purpose is to replace the analog APT (Automatic Picture Transmission) system, still available on NOAA (National Oceanic and Atmospheric Administration) polar satellite fleet. This paper presents some aspects involving the reception and decoding of high-resolution LRPT images provided by the Russian Meteor-M N2 weather satellite, using a VHF reception system based on software-defined radio (SDR). The experimental tests were undertaken in order monitor and evaluate the performance of LRPT broadcast service during in-orbit commissioning test phase of the satellite. This paper also presents some guidelines and conclusions drawn from these experiments.

Keywords: LRPT, SDR, weather, Meteor-M N2.

THE DESIGN AND IMPLEMNTATION OF A PID CONTROLLER IN AN FPGA CIRCUIT

Cristian ORIAN, Dorin PETREUŞ

Abstract: The following paper presents the design and implementation of a PID regulator which is used to control a step-down converter. The transfer functions of both the control block and the buck converter were modeled in Mathcad. These models were instrumental during the design process and results validation, because they were used to plot a step response. The overall feedback loop was simulated in Psim and resulted in the same step response. An FPGA was used for the practical implementation. The circuit was described, synthesized and implemented in Xilinx Vivado. Relevant parts of the control circuit and measurements of the step response are presented.

Keywords: PID control, FPGA, digital control, buck converter.

MICROGRID CONCEPT BASED ON DISTRIBUTED RENEWABLE GENERATORS FOR A GREENHOUSE

Dorin PETREUS, Radu ETZ, Toma PATARAU, Cristian ORIAN

Abstract: This paper presents the modelling and control of a remote all renewable energy microgrid designed to supply a vegetable greenhouse. The microgrid operates in island mode and consists of: a geothermal generator, a biomass generator, a photovoltaic generator, battery storage system and a power management system (PMS). A two level control structure is implemented to properly balance the power flow in the proposed microgrid: primary level based on droop control implemented by each unit and a secondary level base on communication implemented by the PMS.

Keywords: smart microgrid, geothermal energy, biomass energy, solar energy, management.

Number 1/2015

OPTIMIZED SYNTHESIS AND FPGA IMPLEMENTATION OF A FIR FILTER FOR MULTIPLE POSITION EQUALIZATION OF A RECORDING STUDIO/CONCERT HALL

Adrian Călin FĂRCAŞ, Ervin SZOPOS, Ioana SĂRĂCUȚ, Marius NEAG, Marina Dana ȚOPA

Abstract: This paper presents an improved approach to the synthesis and implementation of digital filters employed for multiple position equalization of a room transfer function. First, we propose a more effective way of deriving the desired frequency response of the equalizer than the Levinson-Durbin algorithm employed by well-known method used here as a reference to compare against. Second, the equalizer is implemented by a FIR filter synthesized by using an optimized procedure, based on a genetic algorithm that curtails the differences between the wanted and synthesized frequency responses while minimizing the filter length. These two features help us reduce significantly the hardware requirements for implementing the resulting equalizer. The effectiveness of the proposed method is demonstrated on a real-life example: the implementation of a FIR filter able to equalize the room transfer function of a recording studio used also as a concert hall. The optimized synthesis procedure yielded a FIR equalizer whose number of taps was reduced by an order of magnitude compared to the results reported by the reference method. The circuit implementation of the equalizer required only a small fraction of the resources provided by a mid-range FPGA development board. Objective and subjective measurements were carried out with and without the equalizer being inserted in the acoustic chain; they showed clearly that the proposed equalizer improved significantly the audition.

Keywords multiple position equalization, room transfer function, genetic algorithm, recording studio, concert hall, FPGA

AidISO – THE FIRST STEP FOR SMALL AND MEDIUM-SIZED COMPANIES IN BECOMING ISO 9001:2008 CERTIFIED

Emilia SIPOS, Luiza BUZDUGAN, Laura IVANCIU

Abstract: The standard has become the main tool for separating the product, the reference object of contracts and transactions, the scale of price formation, quality guarantee and product promotion. The theme of this paper is to develop a stand-alone application to assist small and medium-sized organizations in implementing ISO 9001:2008. The software product can be used by companies from five activity fields and it provides time planning for the implementation process, completing the documents, continuous update of the implementation states, saving the changes and the possibility to choose among several Romanian ISO certification bodies.

Keywords: ISO, quality management, Matlab, software tool.

FULLY DIFFERENTIAL CURRENT MODE LOW-PASS BIQUAD WITH INDEPENDENT PARAMETER TUNING

Robert GROZA , Claudia FARAGO

Abstract: The paper presents a fully differential log-domain low-pass second order filter. The topology was derived, using a LINELIN transformation, from a standard state-variable biquad. The filter allow for orthogonal parameter tuning. Its low-frequency gain (H0), natural pulsation (ω0) and quality factor (Q) can be tuned continuously by varying only DC bias currents. The filter was implemented in a generic BiCMOS 0.18μm process and the simulation results demonstrate the validity of the design.

Keywords: ELIN, current mode circuits, parameter tuning

THERMAL INFLUENCES ON ELECTRONIC COMPONENTS FROM THE MANUFACTURING STAGE TO SYSTEM INTEGRATION

Alexandra FODOR, Rajmond JANO

Abstract: The present work investigates the process of thermal management for individual components used for electronic systems, from their manufacturing stage to the assembly of the final product. Two stages of development were studied, firstly, the manual soldering process was investigated and how the thermal phenomena can affect different ICs, depending on their form, during soldering and secondly the effect of the IC package on heat dissipation was investigated. The manual soldering process was studied for the smallest of the surface-mount devices because the assembling technique involves direct contact with the soldering iron. DIP- 28, TQFP-32 and MLF-32 package models have been implemented in SolidWorks 2013 and their efficiency in evacuating heat has been evaluated and compared. Natural cooling was investigated in an open environment and investigations in closed spaces and with forced cooling applied were also made. The research focuses on accurately modelling electronic packages for ICs.

Keywords: thermal, model, soldering, SolidWorks, packages

AN EVOLUTIONARY FPAA RECONFIGURATION ALGORITHM BASED

ON THE OPEN TRAVELING SALESMAN PROBLEM

Claudia FARAGÓ, Paul FARAGÓ, Sorin HINTEA, Gabriel OLTEAN

Abstract: This work proposes an FPAA reconfiguration algorithm formulated on a modified version of the open traveling salesman problem (OTSP), which targets the achievement of a figure of merit (FOM). Genetic algorithms (GA) were chosen to optimize the modified OTSP. Several modifications were made to the genetic operators to guarantee population validity along the evolutionary process. The efficiency of the modified GA is demonstrated over the classical GA for small-sized FPAA mapping, while the classical GA is used for larger FPAAs. Simulation results illustrate the applicability of the proposed evolutionary FPAA mapping algorithm based on the modified OTSP routine.

Keywords: Field programmable analog array, analog filter, open traveling salesman problem, genetic algorithms

ENTERPRISE BUS AS A SERVICE – AN IDEAL CLOUD CONNECTIVITY MODEL

Sorin POPA, Mircea-Florin VAIDA

Abstract: Cloud computing is unquestionably one of the most important things happening in IT today. It’s a concept meant to manage resources in a more responsible manner and to offer companies a professional, efficient way of accessing them. Nevertheless, one of the primary struggles of the cloud-oriented world is application integration. The primary goal of this project is to define an ideal connectivity model for moving the application integration from the organization side into the cloud. In this way, the organization will efficiently save resources: less deployment time, fast and clean new releases, no redundant infrastructure maintaining costs and of course higher productivity. As a lightweight and flexible solution, the “Enterprise Bus as a Service” architectural pattern, or short EBaaS, will manage everything needed by the organization to properly integrate their system into a one-click cloud application.

Keywords: integration, connectivity model, cloud, ERP, EBaaS, ESB.

Number 4/2014

SECURITY ANALYSIS OF THE ARCHITECTURAL COMPONENTS WITHIN A WEB-BASED REPORTING APPLICATION

Dan Bogdan ZAH , Bogdan MIHAILESCU, Andrei Bogdan RUS , Kurt EDER, Virgil DOBROTA

Abstract: In the work presented herein we conducted a security audit of the architectural components included in any web-based data reporting application. Overall we implemented a total of eleven attacks: four of them targeted the server, four the web application on the client device, two envisaged the wireless router and one exploited the vulnerabilities of the database server. Along with the successful Denial of Service attacks, we managed to extract confidential information (i.e. password hashes) from the system using sniffing, Structured Query Language injection and Cross-Site Scripting attacks. For any successful security breaking presented herein, we proposed a set of techniques that can mitigate the specific vulnerability.

Keywords: Denial of Service, Security, web-based application

A DIFFUSION APPROACH FOR ORIENTED VOLUME DATA DENOISING

Romulus TEREBES

Abstract: This paper proposes a new method for denoising non-time dependent volumetric oriented data blocks. The method is developed under the partial differential equations theoretical framework and it is defined on orthogonal section planes of the threedimensional space. The efficiency of the method in denoising oriented volume data is proven using an extensive experimental part involving several random computer-generated synthetic data blocks and statistical interpretations. In the experimental section we also provide a result obtained on real data.

Keywords: diffusion equations, partial differential equations, image restoration

SURVEY OF MR IMAGE PROCESSING METHODS

László LEFKOVITS, Szidónia LEFKOVITS, Mircea Florin VAIDA

Abstract: In this paper a synthesis of the main concepts and the most important MRI image acquisition and processing methods are presented. These two processes are clearly delimited, without any interference in our research means. The magnetic resonance acquisition methods are the exclusive tasks of medical equipment provider companies. The image processing methods –inhomogeneity correction, segmentation, registration and digital atlases – fall on the software developers. The starting points of such a system are the multimodal MR images. The purpose of these methods is the precise delimitation of anatomic structures and determination of benign and malign tissues. The goal of the research is to offer to medical experts a useful knowledge to combine the methods optimally and to obtain better medical results in assisted diagnosis.

Keywords: MRI, inhomogeneity, segmentation, registration, digital atlases.

THE TRANSITION PERIOD OF THE CHARGING OF ACCUMULATORS

Szabó LEVENTE

Abstract: The states of electrochemical phenomena in the accumulators are passing through several phases between the moment of commutation and the permanent period of charging. These successive states appear once at every commutation. This transitory period is lasting approximately 1 millisecond in the case of lead-acid accumulators.

Keywords: transition period, charging of accumulators, Lead-Acid-Battery

APPLYING PRINCIPAL COMPONENT ANALYSIS ON CALL DETAIL RECORDS

Ionut B. BRANDUSOIU, Gavril TODEREAN

Abstract: In the mobile telecommunication industry, the call detail records could have millions of records and thousands of variables. Thus, it is very unlikely that all of the variables are independent, without presenting any correlation. The presence of correlated variables leads to instability in the solution space, thus obtaining incoherent results when implementing a predictive model. In this paper, we apply a principal component analysis on a call detail records dataset to reduce the dimensionality of the data and to eliminate the problem of multicollinearity.

Keywords: principal component analysis, call detail records, reduction of dimensionality, multicollinearity, data mining,

telecommunications.

HUMAN AUDIOGRAM FIR MODELING WITH MATLAB TOOLKITS

Ioana SĂRĂCUŢ, Erwin SZOPOS, Marius NEAG, Călin FĂRCAŞ, Marina Dana ŢOPA

Abstract: The paper presents the synthesis of FIR filters with the magnitude characteristic that matches an arbitrary human audiogram, by using two of the toolkits provided by Matlab: the levinson function, that implements the Levinson-Durbin algorithm, and the standard routine fir2, based on the frequency sampling method. The performances of the filters synthesized with the two Matlab toolkits are evaluated by comparing their magnitude characteristics against the given audiogram, for the same length of the filters. Moreover, for each of the two implemented filters having the optimal length, for the same maximum approximation error, the impulse response and phase characteristic are also discussed.

Keywords: FIR filter, Levinson-Durbin algorithm, human audiogram

COLORIMETER FOR FRUIT SORTING SYSTEM

Lorant SZOLGA, Bianca VLAICU, Ramona GALATUS, Emil VOICULESCU

Abstract: The present paper proposes a study, followed up by the implementation of a low cost and low power consumption fruit sorting system. The implemented system was tested for apple sorting. Under the tests we gained an accuracy of 100% from the correct sorting point of view, which emphasize the good functioning of the system proposed by us.

Keywords: colorimetric, microcontroller, RGB LED, photoresistor

Number 3/2014

A LIFE DEDICATED TO SCIENCE AND INNOVATION – DOCTOR HONORIS CAUSA FOR DR. BARRIE GILBERT

Acta

Abstract: Technical University of Cluj-Napoca awarded the title of Doctor Honoris Causa to one of the greatest pioneers in the field of semiconductor circuits - Mr. Barrie Gilbert, North West Lab, USA. Tuesday, September 2, 2014, the Auditorium Alexandru Domsa, in Technical University of Cluj-Napoca, Romania, hosted an event highly important to the academic community and users of modern data processing and communication devices: the granting of the Doctor Honoris Causa title to Mr. Barrie Gilbert from North West Lab, USA, one of the great pioneers in the field of semiconductor circuits. Author of over 100 inventions started in the USA but used in various applications worldwide, in communications, medicine, transportation and IT, Mr. Barrie Gilbert is widely recognized as a great visionary in the field of semiconductor devices and analog integrated circuits. Also, Mr. Barrie Gilbert is a well known author and highly cited as special reference. The great scholar and man of the art that is Mr. Barrie Gilbert agreed to become Doctor Honoris Causa of the Technical University of Cluj-Napoca as a recognition of the University’s value and rewarding the achievements in the area of scientific research on international scale, for the members of the Faculty of Electronics, Telecommunications and Information Technology.

Keywords: Barrie Gilbert, Doctor Honoris Causa, Technical University of Cluj-Napoca

IMPROVEMENT OF SOME EFFECTIVE PARAMETERS OF C-SI SOLAR CELL FOR BETTER EFFICIENCY

Santanu MAITY, Chandan Tilak BHUNIA, Partha Pratim SAHU

Abstract: To overcome the Shockely-Queisser efficiency limitations, considerable research into the various solar cell loss parameters such as higher reflectance, higher surface recombination, transmission, low light absorption, high series resistance and low shunt resistance is under way. In this paper, some effective solar cell parameters such as reflectance of light from the cell surface, emitter layer thickness and series resistance are investigated using both experimental and simulation studies.

Keywords: Solar Cell, Texturization, Plasmonic Effect, LIP, Quantum Efficiency

COMPLEX EVENT PROCESSING IN SOCIAL MEDIA

Calin RAILEAN, Monica BORDA, Alexandra MORARU

Abstract: In a world where virtual communities get ahead, social media, supported by web and mobile technologies, represents a valuable source of large amounts of data that can encompass the information shared between individuals related to the traffic conditions. Therefore, a metric that measures the logical link between traffic condition and tweets reported in social media, can be useful in estimating the traffic impact. Accordingly, responsible authorities are provided with a measure of the resources needed for handling them, which can contribute to improving facilities such as public transportation services or traffic estimations. This paper aims to analyze the traffic condition covered in social media, more specifically, in twitter data and to reveal the correlation between traffic conditions and twitter messages. We suggest an automatic method for this association in order to discover its impact and importance. The proposed method is tested on a real world dataset and the results show a positive correlation between the automatic method and a set of manually evaluated associations. The usefulness of this research consists of forecasting public transportation needs and supporting events’ organizers to better arrange their planned activities. The results have real applicability and are filled with consistent interpretations.

Keywords: event processing, traffic correlation, twitter data

DESIGN ILLUSTRATION OF AN EKG ANALOG INTERFACE USING GENETIC ALGORITHMS

Bogdana RÂZA, Alexandra RÂZA, Paul FARAGÓ, Sorin HINTEA

Abstract: Electrocardiography (EKG) systems provide a solution for the recording of the heart activity and are indispensable in patient monitoring. Interfacing the EKG system with the human body is done with analog signal processing and operates towards proper signal conditioning. The EKG analog interface consists of an instrumentation amplifier, a notch filter and a highpass filter. This paper proposes an approach to design the EKG interface using genetic algorithms. Two design strategies are used: coefficient equation for amplifier and notch filter design, and curve fitting for highpass filter design. The validity of the evolutionary design results is demonstrated with transistor-level simulation.

Keywords: EKG analog interface, genetic algorithms, coefficient equation, curve fitting

A TUNNEL-BASED SOLUTION FOR SEAMLESS VERTICAL HANDOVER AND LOAD BALANCING

Iustin Alexandru IVANCIU, Andrei Bogdan RUS, Virgil DOBROTA

Abstract: This paper proposes a multi-tunnel architecture with a smart mobile router offering uninterrupted services for its wireless customers. It simultaneously uses several wireless/mobile access networks to the Internet in order to reach the service continuity gateway located somewhere in a broadband network. This customer-oriented approach offers seamless connectivity through vertical handover and load balancing.

Keywords: Handover, load balancing, tunneling

ON THE AUTOMATED PLANNING AND DESIGN OF SMATV SYSTEMS

Radu ARSINTE

Abstract: The paper presents some theoretical and practical considerations regarding the TV information distribution in local (small and medium) networks, using different technologies and architectures. The SMATV concept is chosen to be presented extensively. The most important design formulae are presented with a software package supporting the network planner to design and optimize the network. A case study is realized, using standard components in SMATV, for a 5 floor building. The study proved that it is possible to design and optimize the entire network, without realizing first a costly experimental setup. It is also possible to run different architectures, optimizing also the costs of the final solution of network.

Keywords: Sat TV, SMATV, Multi-switches, Planning, Design

Number 2/2014

ON ROBUSTNESS OPTIMIZATION BASED ON METAMODELS

Anamaria OROS, Marina TOPA, Marius NEAG, Monica RAFAILA, Georg PELZ

Abstract: Electronic systems are usually required to maintain the value of their parameters within set limits over a wide range of conditions, for large variations of internal or external factors. Given the large number of factors that need be considered it is often un-practical to check all and each combination of factors. New approaches, such as the parameter design, are necessary for designing robust systems, with defined performance for all operational conditions, while keeping the development time and production cost to acceptably low levels. This paper describes two methods for parameter design: the classical, analytical approach, based on underlying assumptions and a new simulation based methodology, focused on real-life cases. The proposed method helps the designer to bring a system response to the target and reduce its sensitivity to varying operating conditions. It is demonstrated on a real-life case: optimized sizing of the external circuitry of a low dropout voltage regulator, with the view of reducing its sensitivity to external factors such as temperature and load current. A three-response optimization was achieved also by applying the method on each of the considered outputs and then intersecting the solutions.

Keywords: Robustness Optimization, Metamodels, Parameter design, Voltage Regulator

LONG RANGE SOUND SOURCE LOCALIZATION EXPERIMENTS

Flaviu Ilie BOB

Abstract: Sound Source Localization (SSL) can be used in various domains, ranging from surveillance and security to medicine or multimedia. Most SSL systems rely on different knowledge about the emitted sound or the environment in order to determine the position of the source. In this paper the author presents the experimental results obtained for an unknown sound source. The distance towards the sound source is determined without prior knowledge about the emitted sound, and the environment is assumed to be noise free, including reverberations. Using a system that has only three microphones placed at 1 meter apart, experiments were conducted at ranges of up to 50 meters, at various angle orientations. Results show that it is possible to measure within an acceptable degree of error the distance towards a sound source situated up to 35 meters away.

Keywords: Sound Source Localization, Time Delays of Arrival, three microphones localization, Unknown acoustic source localization, General Cross-Correlation, Sound Source Ranging

AUTOMATED POWER-VOLTAGE CHARACTERIZATION OF PHOTOVOLTAIC PANELS

Paul. N. MARTARI, Botond S. KIREI, Bogdan I. OLAIOS, Marina D. TOPA

Abstract: This paper contributes to the automated power-voltage characterization of photovoltaic panels. The proposed measurement assemblies can be used in conjunction with “quasi seek” maximum power point tracking algorithms. First, the power vs. voltage characteristics of a solar panel was traced using a common assembly that consists of the solar panel, a trimmer resistor and standard laboratory multimeters. This assembly requires a human operator to switch the trimmer resistor and the measurement intervals on the multimeters, thus not suitable for measurement automation. Next, two measurement assemblies were proposed: (i) a voltage controlled current source used to control the current drawn from the panel; (ii) a voltage controlled voltage reference used to drive the voltage on the panel. It was found in both circuits that the current sense resistor (a low side shunt resistor used for current measurement) must be varied to achieve a wide operation range with respect to the solar illumination, so a programmable resistor was added. Automatic decision over the suitable resistor value was based upon illumination measurement with a usual photoresistor. Finally the circuits were built on a test

board, a virtual instrumentation was set up and the power vs. voltage characteristics of a 2W photovoltaic panel were traced.

Keywords: maximum power point, tracking, photovoltaic cell/panel, measurement automation, voltage controlled current source, voltage controlled voltage reference, current sense resistor, low side shunt

HIL TESTING METHOD ENHANCEMENT BASED ON ALGEBRAIC METHOD PREDICTION

T.E. DACHIN, D. PITICĂ, G. CHINDRIŞ, C. M. NEMEȘ

Abstract: The paper defines and implements a method of verifying if the testing system is suited for being implemented or not. In order to try and force a usability prediction on the whole testing process, we used algebraic calculus applied on a testing system, in order to see the correlation between the used signals, in various system nodes. For being successful in predicting the system stability, algebraic calculus was applied, the eigenvalues calculated, and then the correlation between the mathematical results and the real implemented testing system made. As a result of the work we can state from the beginning of test system implementation, if the system is suited or not for implementation. When deciding, the mathematical results back up the decision.

Keywords: Testing system, Algebraic calculus, Eigenvalues, System stability.

OPENFLOW-BASED IMPLEMENTATION OF A GEARBOX-LIKE ROUTING ALGORITHM SELECTION IN RUNTIME

MirceaValeriu ULINIC, Andrei Bogdan RUS, Virgil DOBROTA

Abstract: This paper aims to prove using an OpenFlow-based testbed that a real gearbox-like routing algorithm selection in runtime is feasible. The proper selection and interworking of flow forwarding schemes for Infrastructure and Service Providers may lead to enhanced performances. Limited resources and/or a partial topology are shared by the owner of the network with the service producer, based on a real-time evaluation of the status using the Floyd-Warshall algorithm for single path and the Ford-Fulkerson for multipath routing. At its level, the Service Provider decides the best forwarding scheme involving the Modified Dijkstra’s algorithm and, whenever multipath is needed, the Ford-Fulkerson algorithm. However due to ownership, security issues, and the business models agreed nowadays, only the Infrastructure Providers are actually performing the routing according to own rules (for its users or in case of emergency) and on behalf of Service Providers, according to their rules. Other combinations of algorithms for the existing or new coming types of providers (e.g. for virtual service infrastructure) can be easily tested now in the proposed software-defined networking tool.

Keywords: Floyd-Warshall algorithm, Ford-Fulkerson algorithm, gearbox-like routing algorithm selection, Modified Dijkstra’s algorithm, OpenFlow

Number 1/2014

FUZZY LOGIC CONTROLLER BASED ON AN INDIRECT VECTOR CONTROL OF DUAL STATOR INDUCTION GENERATOR IN WIND ENERGY CONVERSION SYSTEM

Fatima AMEUR, Katia KOUZI, Nachida KASBADJI MERZOUK

Abstract: This paper investigates Fuzzy Logic Controller Based on an Indirect Vector Control of Dual Stator Induction Generator (DSIG)in Wind Energy Conversion System. In first step a field-oriented control of a DSIG is presented. In second step, in order to ensure an optimum operating point and an Maximum Power Point Track (MPPT) giving online a maximum production of electric power for different wind speeds, a conventional PI and then a fuzzy PI speed regulators have been used. Simulation results show clearly the effectiveness and the performance of the suggested fuzzy logic controller.

Keywords: Dual stator Induction generator, Field-Oriented Control, Classical PI controller, Fuzzy logic controller, Wind energy conversion system.

HIGH LINEARITY ANALOG OPTOCOUPLERS – APPLICATIONS INSIDE THE RF SHIELDED ENCLOSURES

Nicolae CRISAN

Abstract: The paper presents several technical issues and hardware implementations to analyze from the perspective of the galvanic isolation the problematic concern of separation between the transceiver currents flow and the PC common ground. Only the information exchange is allowed through two different types of optocouplers: the first allows transceiver control and monitoring and the second, only analog information exchange. In this paper, the effect of the strong RF field over the PC ground, especially under the reactive behavior of the load (impedance mismatching), has been analyzed. The results tends in favor of the new integrated linear optocouplers. These integrated circuits designed for analog signals are using linearization to increase bandwidth. An optocoupler for audio-interface, designed and tested by the author, experiences the best performance in the proximity of the RF field, enclosed in a metallic box together with the RF power amplifier.

Keywords: Computer aided transceiver (CAT),GNU radio, galvanic isolation, EMI Electromagnetic interference

AN OPERATIONAL TRANSCONDUCTANCE AMPLIFIER SIZING METHODOLOGY WITH GENETIC ALGORITHM-BASED OPTIMIZATION

Claudia FARAGÓ, Alexandru LODIN, Robert GROZA

Abstract: This paper proposes a methodology for the parametric design of operational transconductance amplifiers. In the proposed methodology, amplifier sizing is performed in two stages: an equation-based design scheme performs an approximate design, followed by optimization of the design results with genetic algorithms. The feasibility of this method was investigated with extensive simulation. For demonstration purpose, a folded-cascode operational transconductance amplifier was subject to design.

Keywords: : Folded-cascode operational transconductance amplifier, evolutionary circuit design, genetic algorithms

PRE-PROCESSING AND CLUSTERING ALGORITHMS FOR MICROARRAY DATA IN CANCER RESEARCH

Ioana ILEA, Monica BORDA

Abstract: The development of microarray technologies makes it possible to understand the molecular basis of human diseases, like cancer. Microarray experiments may be used to simultaneously asses gene expression values for thousands of genes. Along with bioinformatics and data mining techniques, they are important in finding relationships between gene expression values and clinical outcomes. One of these techniques is represented by the clustering analysis. This paper describes a typical workflow for microarray data and different hierarchical clustering algorithms for DNA microarray data, underlining their advantages and disadvantages

Keywords: microarray, gene expression, pre-processing, normalization, hierarchical clustering, cancer

Labview Tool for Designing FIR Filters For Acoustic EqualiZation

Erwin SZOPOS, Călin FĂRCAŞ, Marina ŢOPA, Marius NEAG, Ioana SĂRĂCUŢ

Abstract: The paper presents a tool for optimal synthesis of FIR filters tailored for the equalization of a room acoustic transfer function. The synthesis is optimized for minimizing the hardware resources required to implement the resulting filters, in particular by reducing the length of these filters. The optimization procedure employs a genetic algorithm developed for this application. The LabVIEW implementation of the filter synthesis tool is presented in some detail, along with a real-life example.

Keywords: FIR filters, arbitrary magnitude response, acoustic equalization, optimization, genetic algorithm

Number 4/2013

PREDICTION OF WAVEFORMS UNDER THE VARIATION OF INPUT PARAMETERS USING NEURAL NETWORKS

Gabriel OLTEAN, Alex PRODAN, Monica RAFAILĂ, Laura IVANCIU

Abstract: This paper introduces a modeling flow for predicting waveforms as a function of parameters, variables in the system generating the waveforms. In order to achieve this goal, a neural network is involved. The model is developed using early-stage simulation data from the automotive industry. Usually, a large amount of data is necessary in order to properly create such a model and successfully train a neural network, and this can be problematic. To address this issue, a model which can be trained with just a handful of characteristics is proposed. The results obtained show that our model can predict waveforms based on input factors with high accuracy.

Keywords: waveform prediction, neural network, modeling, automotive industry waveforms

SECURE MOBILE-CLOUD FRAMEWORK – IMPLEMENTATION ON THE MOBILE DEVICE

D. POPA , K. BOUDAOUD , M. BORDA

Abstract: Secure Mobile-Cloud is a framework proposed to secure the data transmitted between the components of a mobile cloud application. In addition, the framework, takes into account the following aspects: 1) the users options regarding the security level required for private data and 2) the device energy consumption. The framework includes several distributed components. Some of these components are deployed on the mobile device and some of them in Cloud. This paper is focused on the implementation of the Secure Mobile-Cloud framework components on the mobile device. A proof of concept Android prototype is proposed.

Keywords: Mobile Cloud Computing, Applications, Security

A RESOURCE CONSTRAINED PROJECT SCHEDULING MODEL BASED ON JOB PROFILE SCHEMES

Aurelia CIUPE, Ovidiu ARION, Marcel CREMENE, Bogdan ORZA

Abstract: The aim of the current research is to propose a Resource Constrained Project Scheduling Problem extension. Based on an employee profile derived from a job evaluation scheme and mental performance modeling, a set of new constraints and their influence on project tasks have been designed. An objective function of project makespan minimization through buffer management and human cognitive profile has been defined. Several classic prioritization rules have been reconsidered in order to offer an adapted person-centered solution rather than a process-centered one. By applying the developed model to the scheduling process, the total duration for each task has been minimized according to the employee profile. We provide an overview of the proposed model and the developed application used for obtaining computational results, along with a comparison with several commercial related tools.

Keywords: : buffer management, project scheduling, prioritization rules, project makespan minimization, employee profile, intelligent systems, operations management

DEVELOPMENT AND PERFORMANCE EVALUATION OF A TERNARY FUNCTIONS MINIMIZATION TOOL, IMPLEMENTED USING MATLAB

Emilia SIPOS, Lucian Ovidiu CICA, Costin MIRON, Laura IVANCIU

Abstract: This paper presents a tool designed to minimize both completely and incompletely specified ternary functions with two, three and four input variables. The tool is tested using 33 ternary functions and performance such as minimization time and number of output terms is evaluated. One main advantage of ternary system over binary ones is the reduction of chip area required to implement logic functions. Using a minimization process for ternary functions, the required area is even smaller. For the ternary function minimization tool implemented using Matlab, the minimization process is entirely ternary and the ternary-to-binary and binary-to-ternary conversions used in well know minimization software for ternary functions, such as MVSIS and Espresso MV, are eliminated. For tested functions, the minimization time increases with one second, for every new input variable and the maximum minimization time was 2.9 seconds. The smallest number of output terms is obtained in 40% of the cases for one expression of minimized function, in 42% of the cases for the other expression and in 6% of the cases; the number of output terms is the same for both expressions.

Keywords: minimization tool, ternary functions, performance evaluation, Matlab

PARAMETER DISTRIBUTION IN RECEIVER CHAINS: A HYBRID FUZZY-GENETIC ALGORITHM APPROACH

Laura-Nicoleta IVANCIU, Gabriel OLTEAN

Abstract: The paper presents a novel, hybrid approach to parameter distribution in receiver chains, using fuzzy sets and Genetic Algorithms. The method works with global values for gain, noise figure and intercept points, and aims to correctly assign distributed values of these metrics for each block of the receiver chain. Z-shaped and S-shaped membership functions are used for each of the 4 metrics. The objective function of the Genetic Algorithm is defined as a sum of membership degrees. Tests using a zero-IF receiver architecture with 5 blocks, and WCDMA/WLAN specifications, show that the proposed method rapidly assigns correct values for gain, noise figure and intercept points, for each block.

Keywords: parameter distribution, receiver, WCDMA, WLAN, fuzzy, genetic algorithm

Number 3/2013

A NEURAL NETWORK IMPLEMENTATION ON FPGA FOR ECOLOGICAL MONITORING

Diaf YOUSSOUF, Kaddeche MOHAMED

Abstract: In this paper, we present an implementation of an artificial neural network (ANN) for predicting blooms (coloured waters) of a species of toxic phytoplankton called Dinophysis acumunata that prevalent in coastal of Havre (France). A Multi-layer Perceptron (MLP) neural network is designed to predict the rate of algal cells concentration in a short time series. The processed data are extracted from analysis of average numerical values of physical and chemical parameters. The implementation of the designed neural network on FPGA (FPGA: Field Programmable Gate Array), a programmable circuit, help us to set up an automatic monitoring lagoon system.

Keywords: Automated monitoring, Toxic algae, Eutrophication, Phytoplankton, Neural network, FPGA

PREDICTING CHURN IN MOBILE TELECOMMUNICATIONS INDUSTRY

Ionut B. BRANDUSOIU, Gavril TODEREAN

Abstract: In today’s competitive market, the customer retention represents one of the focal concern for telecommunications companies. The telecommunications sector can be considered on the top of the list with an annual churn rate of approximately 30%. Since the cost associated with subscriber acquisition is higher than the cost of subscriber retention, the need for predictive models to identify subscribers that are at risk of churning has become an important aspect for modern telecommunications operators. But, because the pre-paid mobile telephony market is not easily traceable and definable, implementing a predictive model would be a complex process. In this paper, we present an advanced methodology that helps to predict subscribers churn in pre-paid mobile telecommunications industry, by applying data mining algorithms on dataset that contains call details records. To construct the models we will use k-nearest neighbor, logistic regression, and bayesian networks algorithms. To evaluate and compare the models, we use the gain measure

Keywords: bayesian networks, call details records, churn prediction, k-nearest neighbor, logistic regression, telecommunications

LOCAL AND GLOBAL SPECTRAL VISUAL SALIENCY ESTIMATION BASED ON HUMAN VISUAL BEHAVIOR

Oana Loredana BUZATU

Abstract: In this paper a saliency detection algorithm based on local and global information is proposed. The proposed method extends a previous work by calculating the pixels saliency by means of local patches and the global dissimilarities between a patch and all other patches based on color contrast differences and spatial distances. The image patch scale is fixed and in accordance with the size of the regions of interest (ROI), while the spatial difference takes into account the fact that the focus of attention can subtend a certain region from the visual angle. Resulted algorithm is tested in an image retargeting application and proves its effectiveness.

Keywords: local- global visual saliency, human visual sensitivity, image patch size, image retargeting

NOVEL CURRENT LINE SENSOR BASED ON MATCHED OPTOCOUPLERS FOR ACTIVE POWER FACTOR CORRECTION

Paul MARTARI, Dorin PETREUS, Marius NEAG

Abstract: This paper presents a novel line current sensor which allows safe and accurate data acquisition for power factor measurements: it uses a pair of matched optocouplers arranged so that the difference between their outputs is proportional with the line current; thus the second order harmonics caused by the inherent non-linearity of the optocouplers are cancelled. A system for analyzing experimentally the effects the power factor (PF) has over the efficiency of quasi-resonant flyback switch mode power supplies (SMPS) was build based on this sensor. It provides the chronograms of the line voltage and current, along with the most important data related to the power factor and the efficiency of the power conversion. Experimental results shows that active power factor correction improve the efficiency of the SMPS by up to 6%.

Keywords: Line Current Sensor, Power Factor Correction, Quasi-resonant Flyback Converter

COMPARATIVE ANALYSIS OF THREE VERSIONS OF A CAPACITORLESS LDO STRUCTURE THAT SUPPLIES A RAIL-TO-RAIL VCO

Cristian RADUCAN, István KOVÁCS, Paul MARTARI, Marius NEAG

Abstract: This paper presents three circuit implementations of a low-dropout voltage regulator (LDO) structure that supplies a rail-to-rail voltage controlled oscillator (VCO) operating at 2.5GHz. The LDOs comprise only a small decoupling capacitor that can be integrated but they are still able to deal with the large and fast variations of the supply current required by the VCO. For a fair and direct comparison, all three LDO versions use the same pass transistor and are designed to maintain the output voltage ripple at the same level. Two symmetrical OTAs – one optimized for low-power consumption and the other for higher-bandwidth and lower noise – and one folded-cascode OTA optimized for low noise were employed successively to implement the error amplifier within the LDO. The comparison focuses on the effect the LDOs have on the phase noise of the VCO they supply. It was found that the VCO phase noise can be improved by reducing the level of high-frequency noise on the VCO supply line, caused by the LDOs.

Keywords: Capacitor-less LDO, Error amplifiers, SoC

HDRI AND RETINEX TECHNIQUES APPLIED FOR CULTURAL HERITAGE DOMAIN

Oana GUI, Costin MIRON, Petre OGRUTAN

Abstract: The paper presents a comparison between digital images of mural paintings acquired using High Deffinition Range Imaging techniques and images that result after the processing of the images of murals taken under spatial and spectral variations of the illumination with the Retinex algorithm. The processing of mural painting images with the Retinex algorithm elevates their quality to a higher status, comparable with the quality of the images acquired with the HDRI technique. We consider that both techniques, Retinex and HDRI, should become part of the modern products, dedicated for cultural heritage domain.

Keywords: High Deffinition Range Imaging, Retinex algorithm

ON REPLICATION IN DESIGN OF EXPERIMENTS

Bianca FAGARAS, Ingrid KOVACS, Anamaria OROS, Monica RAFAILA, Marina Dana TOPA, Manuel HARRANT

Abstract: Modern electronic technology brings systems of higher complexity leading to the need of new analysis techniques. Using Design of Experiment (DoE) and one of its main principles, i.e. replication, we applied controlled variations in input factors, and replicate the measurements sufficient amount of times, in order to assess the impact of noise on the measurement results, as compared to the impact of the factors. The paper illustrates the replication approach with the measurements performed on a lowpass filter. The method proposed in this paper consists of 4 main steps, which assess the impact of noise on the measurement results as compared to the impact of the factors. Of great interest is determining how noise coming from testing the system under real stimuli, interferes with the results obtained after performing controlled experiments in a simulation environment

Keywords: Design of Experiments, Replication, Low-Pass Filter, Real-Time Load Emulation

BLOCKING IN DESIGN OF EXPERIMENTS

Ingrid KOVACS, Bianca FAGARAS, Anamaria OROS, Marina Dana TOPA, Monica RAFAILA, Manuel HARRANT

Abstract: Real life electronic systems must perform within tight limits even if affected by uncertainties such as temperature and supply variations. Design of Experiments (DoE) plays an important role in improving the quality of a system. The present paper focuses on a robust design method based on one of the main principles governing DoE: blocking. The method is automated with the help of MATLAB and involves the following steps: classify the factors, design the experiment, measure, analyze and optimize the response. The method is illustrated on a low-pass filter. The experiment uses a saturated D-optimal design. The metamodel is obtained considering two noise factors, a blocking factor, a small number of runs and replications.

Keywords: Design of Experiments, Blocking, Randomization, Low-Pass Filter, Real-Time Load Emulation

Number 2/2013

INTERACTIVE TOOLS FOR LEARNING FOREIGN LANGUAGES – CASE STUDY: CLUJ-NAPOCA GERMAN CULTURAL CENTRE

Victor-Eugen SALCA, Roxana-Dorina MOLDOVAN, Bogdan ORZA, Aurel VLAICU

Abstract: The Cluj-Napoca German Cultural Centre organises German language courses for different levels of linguistic competence. In teaching German, the tutors implement interactive learning concepts, mostly serious game. A feedback survey focused on the students’ perception on the interactive techniques used and how further implementing interactive learning techniques improve the German courses was conducted. This paper shows the results of the survey, and interprets them from the perspective of the lifelong learning concept implemented at the Cluj-Napoca German Cultural Centre. The results interpretation was considered when implementing an Adobe Flash® serious game for writing and vocabulary exercises.

Keywords: e-learning, lifelong learning, foreign languages, special education, serious game, problem based learning

ACOUSTIC ECHO CANCELLATION USING WAVELET TRANSFORM AND ADAPTIVE FILTERS

Bianca Alexandra FAGARAS, Cristian CONTAN, Marina Dana TOPA

Abstract: The paper proposes a new method for acoustic echo cancellation(AEC), in order to improve the performances of the normalized least-mean-square (NLMS) and non-parametric variable-step-size-NLMS (NPVSS-NLMS) algorithms. The acoustic system is modeled using an impulse response measured in a low reverberant enclosure. After using the two algorithms for identifying the mentioned acoustic system, one can observe that the NPVSS-NLMS provides better performances than the NLMS in terms of convergence rate at the same steady-state error. Simulations have been performed using both white Gaussian noise and a non-stationary audio signal as source sequences. For further convergence improvement, we propose the wavelet transform for the input signal decomposition.The two adaptive algorithms are applied to the wavelet structure obtained. The proposedwavelet based identification methods provide better convergence rates than each of the two conventional adaptive filters. Also one can notice that the wavelet NPVSS-NLMS(WNPVSS-NLMS) provides

better performances than the wavelet NLMS (WNLMS). MATLAB simulations performed for white Gaussian noise and nonstationary

audio signal used as input prove the previous statements.

Keywords: wavelet transform, adaptive filters, acoustic echo cancellation, convergence rate

IMPLEMENTATION ISSUES FOR A VIDEO STREAMING SERVER IN IEEE 802.11E WLANS

Gabriel LAZAR Tudor Mihai BLAGA Virgil DOBROTA

Abstract: Recent years have seen a proliferation of real-time multimedia traffic over a more and more heterogeneous Internet. Video streaming at high, consistent quality over wireless links proves to be a difficult task. Several optimization techniques have been proposed and studied, mostly through theoretical analysis and simulation. This article describes the implementation of a cross-layer H.264 video streaming server and the evaluation of its performance in IEEE 802.11e WLANs. Measurements present the benefits of employing several key cross-layer mechanisms which aim to improve the video transmission quality over wireless links. A cross-layer signaling solution is implemented, which addresses important QoS issues between user-space and the kernel traffic control and device driver components. Network support for efficient multiqueue transmission is enabled in the Linux network driver. In addition, the paper discusses the implementation of an enhanced scheduling scheme for the receive-side, in order to provide prioritization of video streams over elastic traffic, and also for better control of latency and improved throughput for multimedia flows.

Keywords: cross-layer optimization, H.264, IEEE 802.11e, multiqueue hardware, quality of service, video streaming

PERSONALIZED SECURITY MECHANISM FOR MOBILE CLOUD APPLICATIONS

D. POPA, K. BOUDAOUD, M. CREMENE, M. BORDA

Abstract: Mobile Cloud Computing is a model that brings several advantages to the applications developed for mobile devices. In

this paper, we focus on securing data used and processed by component-based mobile cloud applications. We first show some

criteria that may affect the security properties (e.g. integrity, confidentiality and authenticity) applied to a component-based

mobile cloud application data. Then, we describe a security solution that takes into account these criteria.

Keywords: Mobile Cloud Computing, Applications, Security

AN EFFICIENT APPROACH TO MINIMIZE POWER AND AREA IN CARRY SELECT ADDER USING BINARY TO EXCESS ONE CONVERTER

K. RAMAMOORTHY, T. CHELLADURAI, V. MANIKANDAN

Abstract: Carry Select Adder (CSLA) is one of the fastest adders used in many data-processing processors to perform fast arithmetic functions. From the structure of the CSLA, it is clear that there is scope for reducing the area and power consumption in the CSLA. This work uses a simple and efficient gate-level modification to significantly reduce the area and power of the CSLA. Based on this modification 8, 16, 32, and 64-b square-root CSLA (SQRT CSLA) architecture have been developed and compared with the regular SQRT CSLA architecture. This work evaluates the performance of the proposed designs in terms of delay, area, power, and their products by hand with logical effort and through custom design and layout in 0.18- nm CMOS process technology. The results analysis shows that the proposed CSLA structure is better than the regular SQRT CSLA. A simple approach is proposed in this paper to reduce the area and power of SQRT CSLA architecture. The compared results show that the modified SQRT CSLA has a slightly larger delay (only 3.76%), but the area

and power of the 64-b modified SQRT CSLA are significantly reduced by 17.4% and 15.4% respectively. The power-delay product and also the area-delay product of the proposed design show a decrease for 16, 32, and 64-b sizes which indicates the success of the method and not a mere trade-off of delay for power and area.

Keywords: Application-Specific Integrated Circuit (ASIC), area-efficient, CSLA, low power

NOAA SIGNAL DECODING AND IMAGE PROCESSING USING GNU-RADIO

Nicolae CRISAN, Ligia CREMENE

Abstract: The paper presents several technical issues and software implementations proposed in order to decode, process, and render images received from NOAA satellites. NOAA satellites are seeking extreme weather events on Earth. The proposed solutions offer the possibility of decoding and processing the NOAA signals, by using the PC and its audio-in port, without any extra signal processor-based boards. Also, system architecture is open, according to the GNU requirements and hardware independent, due to its capability of working offline, by decoding a pre-recorded .wav file. Finally, the results are converted to a standard image file that lends itself easily to processing for relevant feature extraction. In spite of its hardware independency, the algorithm is capable to operate also online with SDR receivers. These receivers must follow the Universal Software Radio Peripheral specifications on USB ports.

Keywords: Software Defined Radio (SDR), Universal Software Radio Peripheral (USRP), GNU radio

A NEW APPROACH TO SIGNAL REFLECTIONS ANALYSIS ON PCB TRANSMISSION LINES TERMINATED ON NONLINEAR IMPEDANCES

Monica ZOLOG

Abstract: A novel approach for analyzing signal reflections on transmission lines terminated on non-linear impedances is presented. The starting point of this new method is the Bergeron diagram method. The results obtained by the well-known method are used as input data for the new approach. This new method is implemented in MatLab and is suitable for both linear and non-linear impedances, and also for driving signals having rise/fall times of any value. This is the novelty of this approach compared to the Bergeron method which is valid only for fast command signals. The results are compared and they show good agreement with the ones obtained by experimental measurements performed with a high-performance digital oscilloscope.

Keywords: signal reflections, transmission line, non-linear impedance, time domain measurements

ADVANCED METHODS AND TOOLS FOR ONLINE EVALUATION OF MULTIPLEXING SERVICES AND ENCODING PARAMETERS IN DIGITAL VIDEO BROADCASTING

Alexandru Florin ANTONE, Radu ARSINTE

Abstract: The paper contains a practical study regarding some aspects of multiplexing and encoding implementation in DVB systems. Starting from the ITU-R DTV model, the main blocks are investigated from theoretical point of view, emphasizing the points where is still possible to bring improvements in optimized practical implementation. The theoretical considerations are followed by an experimental study, performed online, using as a core element TSReader Pro, a software package developed to realize the analysis of different aspects (services, multiplexing, RF channel) in DTV (Digital TV). The results are useful in future service improvements, revealing, for example, the unused capacity in each distribution scenario.

Keywords: ITU-R DTV model, DVB, Statistical multiplexing, Multiplex analysis

Number 1/2013

A DERIVED ROBUST STATISTICS APPROACH FOR ADAPTIVE

VOLTERRA FILTERS APPLIED IN NONLINEAR ACOUSTIC ECHO CANCELLATION SCENARIOS

Cristian CONTAN, Marina Dana TOPA, Botond Sandor KIREI, Ingrid Maria KOVACS

Abstract: The paper proposes a novel updating concept for adaptive Volterra kernels that relies on a robust statistics approach. The optimization of a certain cost function leads to the update equations for the linear and quadratic kernels with respect to an optimum error threshold, set according to the relative local noise power. Thereby, for absolute error samples larger than the threshold, which occur mainly in the convergence stage of the adaptive filter, the adaptation is achieved using a sign version of the Normalized Least-Mean-Square algorithm. In the saturation stage of the filter, when most of the absolute values of the error samples are smaller than the particular threshold, the adaptation is carried out based on the Normalized Least-Mean-Square algorithm. This technique tends to eliminate the influence of large valued outliers by improving the convergence of the conventional Volterra filters and maintaining the same misadjustment in acoustic echo cancellation setups. The acoustic system is designed using measured Volterra kernels. The method is tested in terms of Echo Return Loss Enhancement for different probability density functions of the source signal. Also, for the white Gaussian noise case as excitation, a new convergence attribute is introduced for a more precise comparison of the adaptive methods.

Keywords: Acoustic echo cancellation, Volterra filters, adaptive algorithms, robust statistics

CONVERGENCE RATE AND STEADY-STATE ERROR IMPROVEMENT IN

ACOUSTIC SYSTEM IDENTIFICATION USING THE COMBINATION OF LINEAR NLMS ADAPTIVE FILTERS

Ingrid Maria KOVACS, Cristian CONTAN, Marina TOPA

Abstract: The paper proposes the combination method applied on adaptive linear filters in acoustic system identification. The filters are based on the Normalized Least-Mean-Square algorithm and the combination is applied on two filters of different step sizes and different filter lengths. To evaluate the proposed technique, two types of source signals are used: a white Gaussian noise and a non-stationary audio signal. The performance of the adaptive scheme is analyzed in terms of Echo Return Loss Enhancement. The efficiency of the proposed concept was compared with the one of the two sub-band decomposition model. The advantages of the proposed method were outlined in terms of convergence rate and steady-state error.

Keywords: Acoustic system identification, adaptive filtering, adaptive combination of filters, sub-band decomposition

SIMULATION OF DWDM COMMUNICATION SYSTEM AT 8.10 Gbit/s USING CONTINUUM SOURCE IN THE TRANSMITTER

L. GRAINI, K. SAOUCHI

Abstract: In this paper, the 8-channel DWDM communication system at 8.10 Gbit/s is demonstrated. Laser diodes used in conventional DWDM link have been replaced by continuum covering C-band of optical communication window. If the initial source is delivered at repetition rate of 10 GHz, The spectral slicing of this continuum allows obtaining the 8- channel at different wavelengths where each channel is a pulses train has the same repetition rate. Then, they were modulated, in order to achieve data transfer rate of 10 Gbit/s. The modulation format RZ-OOK was used for channels coding. The proposed system components were simulated using COMSIS software.

Keywords: DWDM system, nonlinear optics, continuum, multiplexing, demultiplexing.

RELIABLE POWER INTEGRITY ANALYSIS USING MATLAB

Raul T. FIZESAN

Abstract: In Power Distribution Network (PDN) design of high-speed systems appears to be contradictory design philosophies, component selection and layout rules. A not properly designed power distribution increases the supply rail noise, but this impacts the system’s performance in statistical manners, convoluted with many other variables. For this reason, the analysis and the design of a PDN is crucial in the real world of high-speed board systems. For a good reliability, the PDN should meet the target impedance ([2][5][9][10][13][16]) across the frequency range, up to the highest frequency of interest. Primarily, we need to characterize the high-frequency resonance peaks. Several commercial tools are available to analyze a PDN, [2][14][16][18], but the goal of this paper is to analyze a PDN using a tool developed in MATLAB that calculates and plots the self-impedance of an integrated circuit and the transfer impedance between two integrated circuits of a PCB consisting of a power-ground planes pair. The MATLAB tool gives the possibility to analyze the effects of the planes’ parameters on the noise and to find a tested decoupling strategy to optimize

the PDN.

Keywords: power integrity, decoupling capacitors, target impedance, self-impedance, MATLAB tool

DETECTING DEPENDENCIES BETWEEN STATES OF MULTIPLE DATA STREAMS

Stefan S. DRAGOS, Mircea F. VAIDA, Cristian M. URECHE, Loreta A. SUTA, Alin VOINA

Abstract: The subject of dependency detection has been addressed many times during the last 20 years, and methods that extract frequent patterns or temporal information from data have been developed. The information extracted has been used in fields like medicine, network surveillance, stock analysis and many others, thus proving its importance in real life. But as technology evolved, the amount of generated information grew to the point where storing it for later analysis was no longer an option. In this paper we propose a method of temporal dependency detection between states of heterogeneous data streams, that processes the information as it arrives and discards it afterwards, and we test it on real surveillance data.

Keywords: Data mining, stream mining, temporal dependency detection

ON THE CHOICE OF THE METHOD FOR OBTAINING THE ROOM IMPULSE RESPONSE

Adrian Calin FARCAS, Marina Dana TOPA

Abstract: The paper presents an overview on the direct or indirect methods used to obtain the acoustic room impulse responses. The analysis of the most relevant methods is based on experimental measurements and the goal is to achieve measurements for multiple position acoustic analysis in a live performance room. The objective acoustic parameters of an enclosed space may be obtained from the impulse response of the room.

Keywords: Impulse Response, Maximum Length Sequence, Inverse Repeated Sequence, Time-Stretched Pulses, Logarithmic Sine Sweep

Number 4/2012

DECISIONS TOOLS IN FINANCIAL TIME SERIES PROCESSING: A STOCHASTIC DOMINANCE PERSPECTIVE

Mircea-Cristian GHERMAN, Cyrille PIATECKI*, Monica BORDA

Abstract: The aim of this paper is to provide an implementation of a stochastic dominance algorithm that establish which of the two distributions is preferred by individuals who have an aversive risk profile. Although, this topic related to decision making theory could be applied in almost every filed where one has to choose between empirical distributions, this study is performed on financial markets across Europe. The focus is to emphasis the impact of an important event (or shock) which occurs in the evolution of a time series. In the case of financial market such an event has been considered the latest financial crisis. The relevance of the study consist in that, it offers a different perspective for investors when choosing between the different financial assets. Our approach indicates that using a generalized stochastic dominance concept, the effects of an important event such the financial crisis is caught also in the attitude of individuals in regards to risk. This approach together with the Meyer algorithm for generalized stochastic dominance is a useful tool in risk aversion analysis when choosing among distributions. Its applicability could be also extended to other fields where a distinction between empirical distributions of time series (signals) is necessary.

Keywords: stochastic dominance, utility function, risk aversion, empirical distributions

EVALUATION OF THE PLASMA CLEANED SURFACE OF A HERITAGE WOODEN PAINTING USING IQA METHODS

Oana Loredana BUZATU, Bogdan Tudor GORAS, Liviu GORAS, Emil Ghiocel IOANID

Abstract: In this paper a new approach regarding the assessment of the visual efficiency of the plasma cleaning treatment applied in artwork restoration is discussed. The proposed method, which was tested in the case of a wooden painting, evaluates the competence of the cleaning process using image quality indexes based on structural similarity and metrics based on peak-signal-to-noise ratio (PSNR) combined with human visual system (HVS) properties. A systematic comparison between the effectiveness of these metrics and a discussion concerning their performances in the context of evaluating the plasma treatment were conducted.

Keywords: heritage, plasma cleaning, image quality assessment, structural similarity, PSNR-HVS.

NEURO-FUZZY STEREO CAMERA CALIBRATION ARCHITECTURES USED IN INTERACTIONS WITH MEDICAL VOLUMES

Marius DANCIU, Radu ORGHIDAN, Mihaela GORDAN, Aurel VLAICU

Abstract: Computer assisted diagnosis/surgery systems need improved solutions for visualization and interactions, as the virtual probe tools. This work presents such an interaction tool and its applications in medical imaging. The novelty is represented by the use of neuro-fuzzy systems in 3D virtual probe positioning; two such configurations are presented and compared to the crisp positioning system. We prove that the neuro-fuzzy solutions outperform the crisp solutions in terms of computational complexity and positioning accuracy. The proposed framework integrates several functionalities: positioning of arbitrary cutting planes; 3D editing, measurements and annotations; stereo visualization. Being a flexible architecture, other functionalities could be integrated in the future.

Keywords: neuro-fuzzy, 3D interaction, 3D editing, stereo calibration

ON USING DNA DISTANCES AND CONSENSUS IN REPEATS DETECTION

Petre G. POP

Abstract: Sequence repeats are the simplest form of regularity and analyzing repeats can lead to first clues to evidencing new biological phenomena. Many of the methods for detecting repeated sequences are part of the digital signal processing (DSP) field and, therefore, the numerical representation of genomic signals is very important. Most of these methods use distances and consensus sequences to generate candidate sequences. This paper presents results obtained using a dedicated numerical representation and a mapping algorithm with different distances and consensus types to isolate the position of DNA repeats with specific lengths.

Keywords: genomic signal processing, sequence repeats, DNA numerical representations, DNA distances, DNA consensus

INTERACTIVE LEARNING CONCEPT IN SPECIAL EDUCATIONAL PROGRAMMES FOR HIGH-PERFORMANCE ATHLETES

Victor-Eugen SALCA, Roxana-Dorina MOLDOVAN, Bogdan ORZA, Aurel VLAICU

Abstract: In order for children to maximize their athletic performances as young athletes, they must engage in high-performance sports resulting in the neglection of the academic training. As the athletic career comes to an end, the lack of certain non-athletic competences influences the difficulty of socio-professional reintegration of the former athlete. The training and competition schedules make attending to regular classes impossible. The way the regular playing season and the extra-seasonal training schedule influence the school attendance and academic performance of a cadets level volleyball team is studied. The paper identifies the way e-learning can be used to efficiently help athletes to stay in touch with school, without interfering with their athletic career.

Keywords: e-learning, lifelong learning, distance learning, athletes’ education, Moodle plugin, CAT, virtual class.

AN AFFORDABLE APPROACH FOR EXTENDING THE SDR CONCEPT OVER THE X RADIO BAND

Nicolae CRISAN

Abstract: In this paper several technical solutions are proposed to extend SDR technology to X microwave band, at low cost. This extension requires a hardware combination of modules, most of them readily available on market from commercial suppliers. The great advantage of the proposed system lies on its ability to process digital signals from X band on the PC after several frequency down conversions. To make the X band accessible a set of measures is proposed that aim to transform a complex problem with a high degree of nonlinearity in a simple linear problem by omitting some parameters.

Keywords: Software Defined Radio, Down-converter, X-band, Microwaves

DESIGN OPTIMIZATION OF ANALOG ACTIVE CIRCUITS USING THE GENETIC ALGORITHM

Ioana SARACUT, Marius NEAG, Raul ONET, Istvàn KOVÀCS

Abstract: This paper presents an optimization tool for the design of analog circuits implemented with commercially-available discrete devices. It takes into consideration the standard values available for the passive components, as well as the nonidealities of commercially-available active devices. The multi-criteria optimization is based on a genetic algorithm developed in Matlab specifically for this task, that works in conjunction with components of the OrCAD package, namely the OrCAD CIS Capture and PSpice. The effectiveness of the proposed procedure is demonstrated by a real-life example, the optimized design of a fourth-order low-pass filter implemented with Operational Amplifiers and discrete resistors and capacitors.

Key words: analog circuit design, active filter, genetic algorithm.

AN EXPERIMENTAL ANALYSIS OF SIGNAL REFLECTIONS ON PRINTED CIRCUIT BOARD TRANSMISSION LINES

Monica ZOLOG

Abstract: Signal reflections for several digital integrated circuit families are experimentally analyzed by using high-speed time domain measurements. Four test setups are designed and fabricated - the interconnect lines have different geometrical parameters. The signals that are transmitted on the line are measured using a high-performance digital oscilloscope. Each board is successively equipped with ICs from the analyzed families. Simulation results are presented in comparison with experimental results. It is shown that the circuit model must be in accordance with the manufacturer specifications in order to obtain valid simulation results.

Keywords: signal reflections, microstrip, transmission line, time domain measurement

Number 3/2012

ADVANCED VHF GROUND STATION FOR NOAA WEATHER SATELLITE APT IMAGE RECEPTION

Vlad DASCAL, Paul DOLEA, Octavian CRISTEA, Tudor PALADE

Abstract: This paper presents a short overview of the available LEO weather satellites transmitting APT (Automatic Picture Transmission) images. Also, this paper shows the development and setup of an advanced VHF ground station capable of receiving APT images from LEO weather satellites. This ground station will be capable of automatic tracking of selected NOAA satellites by using a steerable high gain multi-purpose VHF X-Quad Yagi antenna. The 137 MHz FM signal from NOAA satellites is received using a wideband Icom receiver and is processed using an external USB soundcard with dedicated software for APT/WEFAX image decoding. The proposed solution will provide low-cost access to APT data from NOAA POES weather satellite fleet. Tests were completed with good results and based on the experimental data, it demonstrates the technical feasibility of this improved receiving system.

Keywords: Low Earth Orbit, ground station, NOAA, weather, APT

SIMPLIFICATION OF A LINK PERFORMANCE PREDICTION METHOD BASED ON MUTUAL INFORMATION

Mihai-Alin Badiu, Mihály Varga, Vasile Bota, Aurel Vlaicu

Abstract: In this paper, we propose a method to predict the block error rate performance of the wireless links in OFDM communication systems. Our approach relies on a performance prediction methodology that uses the mean mutual information per coded bit as quality measure. We make a simplification within this methodology such that, contrary to existing approaches, our method needs the generation and storage of much less reference performance curves. Simulation results obtained for both additive white Gaussian noise and multipath fading channels demonstrate the very good prediction capability of the proposed method.

Keywords: Link performance prediction, OFDM, block error rate, mutual information, MMIB

IMPROVEMENT OF JOT’S REVERBERATION ALGORITHM

Norbert TOMA,Marina TOPA, Ioana SARACUT

Abstract: The paper considers an artificial reverberation algorithm developed by Jot that simulates the behavior of a real concert hall and takes into account the reverberation time’s variation with respect to frequency by using absorbent filters. This work includes both Jot’s design of the absorbent filters and an improved version of it. Simulations are run and objective acoustic parameters are computed. The just noticeable difference (JND) is used to find the best artificial reverberator to ensure the best acoustic behavior. To assure the best acoustic behavior the just noticeable difference (JND) are used according to ISO standards.

Keywords: artificial reverberation, acoustic parameters, digital filters, just noticeable difference

MODIFIED SETTING NUMERICAL DISTANCE PROTECTION OF POWER

TRANSMISSION LINE IN PRESENCE TCSC USING IEC 62850 COMMUNICATION PROTOCOL

Mohamed ZELLAGUI, Abdelaziz CHAGHI

Abstract: This paper presents a study of the performance of numerical distance relay for a 400 kV electrical transmission line compensated by series Flexible AC Transmission System (FACTS) i.e. Thyristor Controlled Series Capacitor (TCSC) is connected in the middle of the transmission line and can be used in voltage mode control or reactive power injection mode, and damping of power system oscillations in high power transfer levels. IEC 61850 is a new international standard for communication networks and systems in substations that defines different message types based on their transfer time requirements. In IEC 61850 based Substation Automation systems (SAS). The effect of the TCSC on the measured impedance at the relaying point is investigated. The setting zones protection deals with the absence and presence of TCSC using IEC 61850 communication protocol for two stations i.e. Principal station and TCSC station with proposed scheme for tele-protection for studies the effects of TCSC insertion on the total impedance of a transmission line protected by MHO distance relay and the modified setting zone protection in capacitive and inductive mode for three forward zones (Z1, Z2 and Z3) have been investigated in order to prevent circuit breaker

nuisance tripping to improve the performances of distance relay protection.

Keywords: Numerical Distance Relay, Setting Zones, TCSC, Tele-Protection, Substation Automation System, IEC 61850

Communication Protocol

RESONANT CAVITIES FOR DUPLEX FILTERS IN VHF REPEATERS: ANALISYS, IMPLEMENTATION, AND TESTING

Gheorghe SAUCIUC

Abstract: In this paper I presented the role of filters duplex in structure of voice repetors, operating principles specific subassembly transmission and subassembly reception. I wanted to do a filter, for building a repeater duplexer for band 144Mhz - 146Mhz. I described how to make a cavity. Leaving the wiring diagram and I checked, depending on the size, if designed cavity can be used for VHF frequencies corresponding. I examined the response of this filter, for receiving and transmitting frequencies, specific to repetor that will be made.

Keywords: repeater, band pass filters (BPF), band rejection filters (BRF), resonant cavity, attenuation

A NOVEL PREAMBLE DESIGN FOR CHANNEL ESTIMATION IN MIMO OFDM

SYSTEMS RESULTING IN ENHANCED THROUGHPUT

Shakeel Salamat ULLAH

Abstract: Multiple Input Multiple Output (MIMO) system combined with orthogonal frequency division multiplexing (OFDM) provides a reliable solution for enhanced data rate next generation wireless systems as MIMO produces additional parallel channels in spatial domain which allows high data rates to be achieved without extra bandwidth and transmitted power. While OFDM provides high bandwidth efficiency and robustness to multipath fading due to orthogonal subcarriers that convert frequency selective channel into parallel frequency flat channels. Channel estimation is integral part in coherent MIMO-OFDM systems. Burst communication system utilizes data aided channel estimation algorithms due to their fast convergence and accuracy relative to blind estimation algorithms. Data aided schemes in MIMO-OFDM employ periodic transmission of training data from transmitter for channel estimation at the receiver. In MIMO-OFDM systems higher number of transmit/receive antennas causes a bigger overhead for channel estimation. This paper proposes novel preamble design with low overhead for channel estimation. Simulation results and analysis of time based channel estimation algorithm resulting in spectrally efficient MIMO-OFDM system have also been provided.

Keywords: Burst Communication, preamble design, time based channel estimation, MIMO-OFDM

BEARING DEFECTS SIGNALS DEMODULATION USING SHOCK FILTERS

Bogdan BETEA, Mircea-Cristian GHERMAN, Monica BORDA, Petru DOBRA

Abstract: A fundamental problem in the development of faults detection and diagnoses methods is to obtain fault signature data as clear as is possible. This article is focused on vibration signals generated by rolling bearings with localized defects A demodulation method is presented for rolling bearings defects signals, which is used on detection of defects signatures (frequencies). The proposed demodulation methods are based on one dimensional shock filters, which are intensively used on image processing but as two dimensional filters. Different shock filters implementations used for the demodulation method are provided together with a comparison of the results. In order to get a clearer result of the frequency localization for the rolling bearing defects, the shock filter formalism is used together with a Gaussian filtering technique. Then, based on these approaches and on a demodulation method, an identification of the defects is obtained. The obtained results are then compared with a demodulation method which is based on the complex shock filters. The experimental results are in favor of complex shock filter approach, which we considered to be a better approach successfully applied in identification of rolling bearing defects.

Keywords: shock filter, bearing defect, demodulation, envelope detection, vibration

A GENETIC ALGORITHM FOR THE OPTIMIZATION OF THE ANALOG CHANNEL FILTER

Ioana SARACUT, Marius NEAG, Victor POPESCU, Erwin SZOPOS

Abstract: This paper presents a method for optimizing the transfer function of the channel filter within the Zero-Intermediate Frequency radio receivers of Orthogonal Frequency-Division Multiplexing (OFDM) broadcasts. The method is based on a genetic algorithm (GA) and has several distinctive features: the type and order of the transfer function can be changed during the optimization process; the optimal solution is calculated based on three objective functions that allow an independent control of the pass band and stop band; a real test signal including wanted and adjacent channels is applied at the filter’s input, which allows the estimation of the nonlinearity effects of the filter. Another important feature is that for a given set of conditions, the GA is run multiple times, taking full advantage of the random permutations employed by the algorithm: the multiple results are not only used to select the very best but they are also analyzed statistically in order to estimate the convergence of the algorithm.

Keywords: circuit optimization, genetic algorithm, analog channel filter, OFDM radio receiver

MIPIOS – MEDICAL IMAGE PROCESSING APPLICATION ORIENTED FOR MOBILE DEVICES

Marius-Cristian URECHE, Mircea-Florin VAIDA

Abstract: Mobility is thought as a key concept that fundamentally changed our way of working and interacting during professional life. However, mobility is a concept of particular interest in the field of medicine and particularly in tele-health. In this paper we present a novel application whose main objective is to take advantage of the major advances of mobile devices for immediate application in the medical domain. We designed our application entitled MIPiOS (Medical Image Processing on iOS) for manipulating, analyzing and processing medical images in DICOM format by using the benefits of touchscreens. With the addition of the Retina Display for several mobile devices produced by Apple, our proposed application’s benefits are even more easily to be exploited because of better finger gestures recognition, better resolution and

better contrast.

Keywords: medical image processing, iOS-based mobile devices, DICOM image visualization, mobile software application.

Number 2/2012

TEST POWER OPTIMIZATION WITH REORDERING OF GENETIC TEST

VECTORS FOR VLSI CIRCUITS

Balwinder SINGH, Sukhleen Bindra NARANG, Arun KHOSLA

Abstract: Power optimization is one of the important challenges in VLSI circuit for testing engineers. Larger power dissipation becomes the reason for overheating and with every increase in 10oC in operating temperature, failure rates for the component on a chip doubles. Power dissipation is directly proportional to switching activities of the components on Integrated Circuits. Power optimization is possible only by minimizing the toggling count (switching activity) for combinational and sequential components on a chip area. This paper describes novel technique of power optimization by rearranging the test patterns generated by Genetic Algorithm. The logic discussed here calculates and re-arrange these genetic test patterns according to minimum toggling arrangement of test patterns. This algorithm is applied on the ISCAS85 and ISCAS89 benchmark circuits. The experimental results show that maximum power dissipation in the combinational and sequential logic circuits are reduced by the average of 31% and 36% respectively.

Keywords: GA (Genetic Algorithm), power optimization, Hamming Distance (HD), CUT (Circuit under Test)

ROBUST FEATURE SETS FOR THE SPEECH PROCESSING

Zhandos YESSENBAYEV

Abstract: This paper compares the various feature sets for speech signal such as cepstral-based and acoustically-driven parameters to analyze and see how robust they are in different noise environments. Despite the success of the cepstral-based features in the tasks of speech processing, they are still susceptible to noise. In this work, we explored how to extend MFCC-based feature sets with other acoustically-driven parameters to add some robustness for the segmentation of speech. Experiments conducted on TIMIT dataset using the standard HMM/GMM framework show that better performance can be achieved if cepstral-based features are combined with acoustic parameters.

Keywords: MFCC, acoustic parameters, HMM, speech segmentation, TIMIT

CODING TECHNIQUE FOR INFORMATION SENT THROUGH A PCB PARALLEL DATA BUS FOR AVOIDING CROSSTALK

Mihai DARABAN, Dan PITICA

Abstract: This paper presents a crosstalk avoidance code (CAC), developed by analyzing the integrity of the signals sent through a

printed circuit board (PCB) bus. There was also determined the influence of trace impedance dynamic variation regarding the

switching pattern. In CAC are integrated rules related to switching patterns that help creating a forbidden transition channel

(FTC). A FTC has the advantage that does not only reduce the crosstalk effects but also resolves the inter-symbol interference (ISI)

propagation problems. The CAC developed has as purpose an error free transmission at high speeds on a PCB build with low cost

technologies, without using high speeds manufacturing processes.

Keywords: switching pattern, bus coding, crosstalk, characteristic impedance

SIMULATION OF THE FORD–FULKERSON ALGORITHM USING OMNET++

Paula SEVASTIAN, Andrei Bogdan RUS, Virgil DOBROTA

Abstract: This paper presents a contribution to the OMNeT++ 4.1 simulator in C++ by integrating our own implementation of the FF - Ford-Fulkerson multipath routing algorithm. In order to validate it, an eight-node testbed works with dynamic requested and released flows on each link, updated every second. Four simple scenarios were proposed, considering all possible combinations of fixed and/or random variation laws. Additionally, the path search algorithm BFS - Breadth First Search was implemented too. This software package running under Fedora Core gets the input data from three files called Nodes, Connections and Params. This mechanism provides the interworking between the C++-based FF simulator and the Java-based application in Intellij IDEA 10.5, used as traffic variation simulator. The graphical representations of the occupied, the requested and, respectively, the released flows are obtained by aggregating the contributions of all flows from the paths for any given source–destination pair.

Keywords: Breadth First Search, Ford - Fulkerson, multipath routing algorithm, OMNeT++

GLOBAL BLUR ASSESSMENT AND BLURRED REGION DETECTION IN NATURAL IMAGES

Loreta A. ŞUTA, Mircea F. VAIDA

Abstract:In this paper we present a global no-reference blur metric based on a local approach of blur detection in natural macro-like images. The purpose is to study the possibility of global assessment based on thedetection of blurred regions over an analyzed image. In our case, it represents the first step for a plant recognition system. Blur detection works on small non-overlapping blocks using wavelet decomposition and edge classification. At the block level the number of edges is less than on global images. A set of rules is obtained by a supervised decision tree algorithm trained on a manually labeled blurred/un-blurred image blocks which leads to a qualitative decision of the blurriness/sharpness of the regions. Experimental results show how the qualitative decision may be transformed in a global assessment.

Keywords: global blur assessment, local blur detection, no-reference blur metric, wavelet analysis

ACCELERATED AGEING TESTS OF ALUMINUM ELECTROLYTIC CAPACITORS FOR EVALUATING LIFETIME PREDICTION MODELS

Rajmond JÁNÓ, Dan PITICA

Abstract: During the course of the presented work accelerated ageing tests at constant temperature for aluminum electrolytic capacitors were carried out. The obtained results from practical tests are compared with projections of the lifetime of the capacitors given by current theoretical prediction methods and the accuracy of these methods is evaluated. Finally, two prediction methods are combined in order to give a higher accuracy lifetime prediction algorithm.

Keywords: lifetime prediction, aluminum electrolytic capacitors, reliability testing

AN APPROACH TO NONRIGID US-MRI REGISTRATION OF PELVIC ORGANS FOR ENDOMETRIOSIS DIAGNOSIS

Marius-Cristian URECHE, Mircea-Florin VAIDA, Stefan DRAGOS, Loreta SUTA, Alin VOINA

Abstract:In the field of multimodal medical image registration, the main elements that researchers focus on are the transformation, the similarity metric and the optimization algorithm. There is no state-of-art registration method for pelvic organs when endometriosis is taken into consideration. This paper proposes a chain of practical implementations for the aforementioned key elements of the image registration framework so that pelvic organs and endometrial cysts are registered between ultrasound (US) and magnetic resonance imaging (MRI). For this purpose, the Free-Form Deformation (FFD) was taken into consideration as spatial transformation, in conjunction with mutual information as similarity measure and gradient descent as optimization strategy. A multi-scale image registration algorithm is implemented to increase the speed of convergence. Registration results show that our approach is feasible for multimodal medical image registration of pelvic organs.

Keywords: medical image registration, free-form deformation, ultrasound, magnetic resonance imaging, B-spline functions

AUTO-COMPENSATION METHOD FOR A BUCK CONVERTER

Toma PATARAU, Radu ETZ, Dorin PETREUS

Abstract: The paper proposes a method to compute the passive components values of a synchronous buck converter. The method is implemented on a step down converter using a DSC(digital signal controller) F28027 from Texas Instruments. A third order IIR filter is used for compensation and the coefficients are computed automatically based on the components determined values. The algorithms for computing the inductance, capacitance and capacitor parasitic resistance are presented in detail with flow charts and simulations. The software is then transposed in code composer and the algorithm is tested on the experimental setup. The algorithm once implemented doesn’t need recalibration of the compensator coefficients made by an operator in case of a hardware change in the power circuit.

Keywords: auto-compensation, digital control, step down converter

DECOUPLED SECOND-ORDER VOLTERRA FILTER STRUCTURE USING THE NLMF ADAPTIVE ALGORITHM FOR NONLINEAR ACOUSTIC ECHO CANCELLATION

Cristian CONTAN, Marina TOPA

Abstract: The paper proposes a nonlinear acoustic echo cancellation setup based on a decoupled scheme of the second-order adaptive Volterra filter. The acoustic distortions that take place in the Loudspeaker-Enclosure-Microphone setup are modeled using measured linear and quadratic Volterra kernels. The adaptation of the filter weights is achieved by exploiting the Normalized Least-Mean-Fourth algorithm. The proposed method is compared with the usual direct adaptation procedure of the second-order Volterra filter using the same algorithm and the same step size values. Comparisons between these techniques regard the evolution of the Echo Return Loss Enhancement in terms of convergence rate and steady-state error. Experimental results for Laplacian distributed excitation show that the new approach increases the speed of convergence in comparison to the one given by the coupled version without enhancing the steady-state error.

Keywords: Acoustic echo cancellation, Volterra filters, adaptive algorithms, nonlinear distortions

A MATLAB GUI FOR SIMULATING THE PROPAGATION OF THE ELECTROMAGNETIC FIELD IN A 2-D INFINITE SPACE

Ioana SARACUT, Victor POPESCU, Marina Dana TOPA

Abstract: This paper presents a Matlab-based graphical user interface (GUI) that allows the simulation of the propagation of an electromagnetic wave in a 2-D infinite space. The homogeneous, isotropic medium is modeled by using the Finite– Difference Time-Domain technique and its infinity is modeled with the Perfectly Matched Layer. The GUI provides different ways of displaying the graphical results: 3-D animation (or snapshots from the animation) of the wave propagation, field variations in time recorded by probes placed in the medium, the average power spectrum and also calculates the reflection coefficients. The capability of the proposed GUI is demonstrated with a numerical example: the simulation of the wave propagation in a field with cylindrical symmetry.

Keywords: Finite-Difference Time-Domain, Perfectly Matched Layer, Matlab GUI, electromagnetic field.

Number 1/2012

PAIRING FOR ELLIPTIC CURVES WITH EMBEDDING DEGREE 1 DEFINED OVER PRIME ORDER FIELDS

Bogdan TUDOR

Abstract: In the last years, prime order pairing-based cryptography has received much attention. The embedding degree 1 is very appealing because the computations can be performed without any extension fields, but the efficiency parameter and this particular value of the embedding degree imposes some limits about the possible curves. This paper presents an analysis of the existing algorithms for constructing such curves and how they are related to the concept of pairing friendly. The original contribution of this paper consists in the proposal of two new algorithms. Their results are analyzed from the efficiency and computational point of view. In addition, it is presented the results of pairing implementation in Java language for the curves generated with the successful algorithm.

Keywords: : computer software, polynomial methods, data privacy, algorithms, elliptic curves

WAVE ITERATIVE PROCESS STUDY OF DOUBLE DIPOLE SCATTERING

Nicolae LUCANU Irinel Valentin PLETEA Ion BOGDAN Henri BAUDRAND

Abstract: The paper presents an application of the Wave Iterative Process (WIP) in the case of the study of the scattering of an electromagnetic plane wave by a system composed of two arbitrary placed metallic dipoles. The case of one dipole is studied at first, a second one being introduced into the scattering system later. Mutual influence is also taken into account, current density results are presented, and the possibilities of enlarging the number of dipoles in the scattering structure are also examined.

Keywords: Wave Concept, Iterative Process, metallic dipole, scattering

CHARACTERISATION OF SMALL CELLS NETWORKS DEPLOYMENT OPTIONS AND THEIR IMPACT UPON MACRO-CELULLAR NETWORKS

Cristian ANDRONE Tudor PALADE

Abstract: This paper investigates the effect of small cells networks deployment in a typical traditional macro-cellular mobile communications network. The main deployment options are presented and their impact is shown for the case of a typical residential environment. Through simulation and modeling the effect of small cells deployment is studied mainly at the macrocell users site, considering different options for the characteristic radio parameters and deployment density of the small cells. The results show which are the feasible implementation options for the small cells within macrocells, and give essential guidelines for operators when deciding to introduce small cells

Keywords: Small Cells, Femtocells, Interference, Open Access, Closed Access, Coverage

ILLUSTRATION OF AUTOMATIC DIGITAL SYNTHESIS FOR CAMELLIA-128 CIPHER ALGORITHM

Paul FARAGÓ, Thomas JACKUM, Christoph BÖHM, Maximilian HOFER

Abstract: Advances in the integrated circuit (IC) industry have enabled the integration of millions of transistors on a single chip. In these circumstances the task of the circuit designer is ever more demanding. Recent trends in CAD tools target the complete automation of the digital design flow. This article illustrates the automatic synthesis of a digital system which implements a cipher algorithm. The Camellia-128 cipher was chosen for implementation due to its potential performance in terms of IC area, processing speed and power consumption. The digital IC is automatically synthesized up to layout level with no human intervention.

Keywords: digital design automation, CAD, cipher, Camellia

HUMAN VISUAL PERCEPTION CONCEPTS AS MECHANISMS FOR SALIENCY DETECTION

Oana Loredana BUZATU

Abstract: : Predicting human eye fixations is a difficult task that usually involves complex knowledge about the human visual perception pathways. This paper proposes a method to detect salient locations using a psycho visual space based on color perception, perceptual decomposition of visual information in multiple processing channels and contrast sensitivity as basic concepts. Using a linear support vector machine to train a saliency detector directly from human eye fixation data, salient locations were predicted as a linear combination of features. Testing results have proved that the proposed method performs better than other state-of-the-art methods with certain common concepts.

Keywords: visual attention, saliency, color opponency, steerable pyramids, contrast sensitivity, support vector machine

INTEGRATION OF NETWORK CODING TECHNIQUES IN FUTURE INTERNET ARCHITECTURES

Zsuzsanna Ilona KISS, Zsolt Alfred POLGAR, Gavril Ionel MEDAN, Mircea GIURGIU

Abstract: The paper introduces several new methods and experimental results on the integration of Network Coding techniques into the architectures proposed for the future Internet. Also, there are shortly presented some of the main research initiatives proposing the design of future Internet architectures and the concepts developed. Particularly, the paper analyzes how Network Coding can be integrated into service centric and role based architectures and proposes a solution for integration of XOR based coding into the Generic Path framework. It is shown that the coded transmissions are not always beneficial and therefore adaptive activation of Network Coding is required according to the communication context and network parameters

Keywords: Future Internet, Generic Path framework, integration, network architecture, Network Coding

INVESTIGATING METHODS TO SIMULATE AND EVALUATE SIGNAL DISTRIBUTION IN ANALOG AND DIGITAL TV CABLE NETWORKS

Radu ARSINTE, Alexandru Florin ANTONE

Abstract: The paper contains a theoretical and experimental study of transmission in analog and digital TV cable networks. The first part presents an overview of the most important concepts used in modeling of the most important components from cable network, and establish the models used in this specific research. The study is proposing in the first part to simulate, using Matlab models a TV cable network. This model is compared with different approaches, like electrical model simulation using Microcap environment. Finally, in experimental results section, the results of simulations are presented and compared with the evaluations realized by previous methods of simulation. The final goal of the research will be to validate, through cross-checking, the models written in Matlab (for theoretical study of DVB-C propagation in local -home – network) or practical purposes (CATV based data network design).

Keywords: CATV network, Matlab simulation, Microcap , Headend

IMPROVED SPATIAL SCALABILITY FOR VIDEO COMPRESSION

Vipan KAKKAR

Abstract: Spatial scalability (layering) in MPEG is not often used due to the lack of coding efficiency and available ICs. This work presents an alternative spatial scalability schemes that have better coding efficiency, and yet are relatively easy to implement with existing building blocks. Furthermore it presents experiments that have been conducted in order to improve the coding efficiency, and enable a backwards compatible HDTV DVD system. Attempts have been made to improve the spatial scalability of MPEG-2/MPEG-4.

Keywords: Video Compression, MPEG, Motion Estimation, Spatial Scalability

LOW-COST PROTOTYPE EQUIPMENT FOR VLF RADIO WAVE MONITORING

Paul DOLEA, Vlad DASCAL, Tudor PALADE, Octavian CRISTEA

Abstract: The paper presents a short overview of the natural phenomena that generates VLF, ELF and SLF electromagnetic waves. In order to emphasize the importance of studying these waves, the paper presents a prototype equipment used for receiving and monitoring very low frequency electromagnetic waves generated by natural and artificial sources. The functionality of the equipment is proved by monitoring, during several days, different VLF radio transmitters located at distances over 1000 km from the reception point. The data collected revealed the diurnal variations of the VLF radio wave propagation and the induced disturbances in propagation conditions during January 4th 2011 partial solar eclipse.

Keywords: radio wave propagation, ionosphere, VLF, low noise amplifier

Number 4/2011

AN OVERVIEW OF TODAY’S MULTIPATH ROUTING

Georgeta BOANEA Melinda BARABAS Virgil DOBROTA

Abstract: The increased number of applications with strict demands regarding throughput and delay (e.g. video streaming, VoIP, video conferencing) determines the need for more efficient routing procedures in the Internet. Multipath routing is one solution that can assure a more reliable and robust transmission. This approach provides flexible load balancing and congestion avoidance by offering multiple alternative paths to a destination. This paper presents an overview of the design details of today’s multipath routing solutions, with their advantages and disadvantages.

Keywords: fast rerouting, load balancing, multipath routing, path diversity, split granularity

RESOLUTION IMPROVEMENT OF ULTRASOUND IMAGES USING DECONVOLUTION AND SUPER-RESOLUTION ALGORITHMS

Liviu-Teodor CHIRA

Abstract: Ultrasound imaging is a wide spread technique used to view body soft tissues, like tendons, muscle or internal organs for possible problems. These images are achieved by capturing the returned acoustic waves using an ultrasonic transducer. Due to the propagation in imperfect environment, the resolution of these images has a poor quality. This is caused by different problems like waves refraction, attenuations in tissues, destructive waves, bad contacts, and so on. This paper intends to achieve a preliminary synthesis of the most important directions of study for the resolution improvement. We present two important methods for ultrasound image processing: deconvolution based and super-resolution algorithms. Some simulation results are also delivered.

Keywords: ultrasound, medical image, super-resolution, deconvolution, digital image processing.

FPGA IMPLEMENTATION OF LMS AND NLMS ADAPTIVE FILTERS FOR ACOUSTIC ECHO CANCELLATION

Ioana HOMANA, Irina MURESAN, Marina TOPA, Cristian CONTAN

Abstract: The paper proposes a register transfer level (RTL) description of two well-known adaptive algorithms used in acoustic echo cancellation: the Least Mean Square (LMS) and Normalized Least Mean Square (NLMS). The RTL descriptions of both algorithms are based on their finite state machine (FSM) diagrams and were created in the popular VHDL language. ModelSim simulation results altogether whit plots obtained in Matlab prove the good behavior. As it was expected, the performance of the NLMS algorithm is superior, compared to the LMS algorithm, in term of Mean Square Error (MSE).

Keywords: VHDL description, adaptive filters, LMS algorithm, NLMS algorithm

FINGERPRINT IMAGE ENHANCEMENT USING WAVELET TRANSFORM AND GABOR FILTERING

Vipan KAKKAR, Abhishek SHARMA, T.K. MANGALAM, Pallavi KAR

Abstract: Automatic Fingerprint Identification System (AFIS) have become the most widely and popular technology in rapidly growing bio-identification applications. Fingerprint identification consists of various stages viz. image acquisition, enhancement, feature extraction and matching. The performance of a fingerprint recognition system highly relies on the quality of the fingerprint images. The enhancement is an essential step required to improve the quality of the fingerprint image. In this paper, we propose a enhancement method based on gabor filtering in wavelet domain. Gabor filter is so chosen because it has both frequency-selective and orientation-selective properties and has optimal resolution in both spatial and frequency domain. Filtering is done on the images results from wavelet decomposition and finally, the image is reconstructed to get the enhanced image. Experiments are conducted on 500dpi resolution fingerprint images commercially available from FVC2002 fingerprint database

Keywords: Fingerprint Identification, Image Enhancement, Gabor Filter, Wavelet Transform

STUDY OF EFFECTS OF THE SHORT TIME FOURIER TRANSFORM CONFIGURATION ON EEG SPECTRAL ESTIMATES

Nicolae Cristian PAMPU

Abstract: EEG signals recorded from scalp contain useful information about the activity of a large number of neurons. Signal processing is needed to extract this information from the EEG signal. Here we study the effects of configurations of Short Time Fourier Transform (STFT) to determine how the parameters of STFT affect spectral estimations of the mean and relative power in beta and gamma frequency bands of the EEG signal. A statistic analysis was performed showing the effects of several window types and window lengths. The estimation of power in a specific frequency band is affected by the configuration of the STFT.

Keywords: Electroencephalogram (EEG), Short Time Fourier Transform (STFT), EEG signal processing

STUDY OF THE DECODING COMPLEXITY FOR RATELESS ERASURE CODES

Anghel BOTOŞ, Vasile BOTA, Aurel VLAICU

Abstract: Over the Internet, bit errors within the data packets translate into packet losses at the higher layers of the OSI model, yielding a packet erasure channel. Modern erasure correcting codes promise to offer a very simple and efficient solution to data transfers over these channels, opening up also other interesting applications. Amongst them one can enumerate reliable large scale content distribution, high quality real-time data transfers, distributed storage and others. These considerations make the study of such codes an actual and interesting topic. Most of the analyses presented in literature focus on the evaluation of the performances ensured by these codes. This paper presents an evaluation of the decoding complexity of some rateless erasure codes, which is another relevant issue that affects the applicability of these codes. The complexities of several decoding methods are evaluated using several metrics which reflect different operations performed during the decoding. All results used in the evaluation are based on simulations

Keywords: rateless erasure codes, computational complexity

ASPECTS OF A LOW-COST GROUND STATION DEVELOPMENT FOR GENSO NETWORK

Vlad DASCAL, Paul DOLEA, Tudor PALADE, Octavian CRISTEA

Abstract: This paper presents the development and setup of a GENSO ground station at the Technical University of Cluj-Napoca. The ground station will be capable of automatic tracking of selected LEO satellites by using steerable high gain antennas. The ground station will be equipped with a full-duplex VHF/UHF transceiver with SSB, CW and FM modulation capabilities and with a TNC AX.25 protocol compatible, that offers data rates of 9600 bps FSK and 1200 bps AFSK for digital satellite communication. The architecture is based on the design heritage of the ground station built in support of the future Romanian Goliat mission. Tests were completed with good results and demonstrate the technical feasibility of the platform based on link budget calculations. The ground station will represent the Romanian contribution into the worldwide GENSO network and it will also act as a test-bed for the GEOID/HumSAT system.

Keywords: Nanosatellite, Low Earth Orbit, GENSO network, ground station, Cubesat

A MODEL OF A DIPOLE ANTENNA IN A 3-D FDTD SPACE

Ioana SĂRĂCUŢ, Victor POPESCU

Abstract: In this paper a dipole antenna with a gap is modeled in a 3-D Finite–Difference Time-Domain coordinate space surrounded by Perfectly Matched Layer (PML). Unlike the analytical study, where the sinusoidal distribution of the current is imposed, here the propagation of direct and reflected waves of current and potential was modeled, so distribution along the antenna resulted from simulations. The simulation results include the radiation pattern, the transitory state of the field and the effect of the conductor losses

Keywords: dipole antenna, Finite-Difference Time-Domain, Perfectly Matched Layer, Matlab model

DEFECT DETECTION AND RESTORATION OF CULTURAL HERITAGE IMAGES

Mihaela CISLARIU, Mihaela GORDAN, Aurel VLAICU, Camelia FLOREA, Silvia CIUNGU SUTEU

Abstract: One of the major problems of the old artworks is their degradation due to aging. As a result, the digitized artworks images exhibit a series of defects. Whereas scientific methods were applied to the analysis and restoration of the priceless artworks for many decades, over recent years there has been an encouragement for the development of virtual artworks restoration tools, many of them using image processing techniques. In this paper we present a series of tools for defect detection and virtual restoration of digitized artwork images, and illustrate their application of digitized wood paintings. These tools are integrated in user-friendly semi-automatic software applications. The novel algorithms proposed and integrated in our virtual artwork restoration application include: a machine learning based small defect detection, combined with the vector median filter for the small defects removal; a fuzzy image inpainting method for large defect removal (assuming the defect area was manually selected). The results achieved with the proposed software application are presented, for some wood painted icons, showing the good performance of the proposed solution and its suitability for virtual restoration of some digitized artworks

Keywords: defect detection, SVM, RGB color space

THE IMPLEMENTATION OF SCHROEDER REVERBERATOR ON AN FPGA PLATFORM USING XILINX SYSTEM GENERATOR

Corina BOTA, Botond Sandor KIREI, Albert FAZAKAS, Marina TOPA

Abstract: The paper presents a Simulink model based FPGA implementation method of the Schroeder reverberator. The Simulink model was created with a Xilinx System Generator toolbox, which aids digital circuit synthesis for FPGA platforms. The artificial reverberator can be configured to simulate the acoustics of different enclosures (ordinary room or concert hall), while device utilization on the target FPGA is low and less than 3% of a XC5VLX110T FPGA device slices are occupied by the designed system. The reverberator can model up to 2s reverberation time, which is a typical value of a concert hall

Keywords: reverberations, audio signal processing, real-time processing, FPGA, Schroeder reverberator

Number 3/2011

FEASIBILITY OF ULTRA LOW ENERGY DEVICES AND APPLICATIONS

Vipan KAKKAR

Abstract: In this paper the results of an extensive literature study on Ultra-Low-Power (ULP) electronics is presented. It discusses an emerging application area, which is characterised by very strong power requirements, called ULP electronics. It aims to evaluate possible applications to derive the requirements that emerging trend in circuits must meet and to recognise as far as possible the challenges, which have to be faced. At first few selected applications which would at most benefit from an ULP design approach, are presented as example. Finally few examples of realized circuits in the market as well as in the literature are given.

Keywords: Ultra-low-power electronics, low-power electronics, biomedical electronics, wireless sensor network

DISTRIBUTED BENDING MEASUREMENT SENSOR

Ramona V. GALATUS, Lorant SZOLGA, Emil VOICULESCU

Abstract: The paper introduces an improved bendable distributed optical fiber used into an all-optical power-up system. The contribution highlights the high sensibility of the large mode area (LMA) fiber, designed with the most powerful higher-order mode rejection. The considered LMA were used in conjunction with MOFA power-up configuration scheme. Setup has a master oscillator fiber amplifier (MOFA) configuration for large scaling of the output power, meant for a linear sensing accuracy of the order of microns. A power variation of the fundamental mode was measured in order to discern the proportional rate for bending radius. The signal was 1550nm, typical to assure compatibility with telecommunication links. The performance is optimized by varying the bending radius, index/ doping profile, the LMA fiber custom profile and core diameter

Keywords: Fiber Optic Sensor, Bending Measurement, EDFA(Erbium Doped Fiber Amplifier), LMA, Highly Doped Fiber, MOFA

GENETIC ALGORITHM FOR TRANSITIONS SCHEDULING GUIDANCE OF TRAINS

Maria-Magdalena SANTA Octavian CUIBUS Tiberiu LETIA

Abstract: The railway traffic system is characterized as large and dynamic with uncertain properties related to resource loading, train arrivals and failures. The purpose of this paper is to control system online in order to guarantee that all the trains behave according to their timelines. The solutions are found with distributed expert systems. The control signals are implemented and verified using temporal Rule Nets. Due to the large number of variants involved, extensive check of all possible durations and waiting times on a resource is impossible, therefore an evolutionary approach (genetic algorithm) must be used. The results obtained through simulations show that the proposed distributed controllers solve the control problems and can be used for large scale implementation.

Keywords: distributed expert systems, genetic algorithm, temporal rule nets and railway applications

3D RECONSTRUCTION AND VOLUME COMPUTING IN MEDICAL IMAGING

Ligia-Domnica CHIOREAN, Teodora SZASZ, Mircea-Florin VAIDA, Alin VOINA

Abstract: Medical imaging domain is a field of interest which concentrates great efforts to offer software tools for the assisted diagnosis. Segmentation, 3D reconstruction and visualization are techniques that allow physicians to observe possible pathological structure inside of the human body. Computing the volume of such a structure offers critical information in the evaluation of the disease gravity. In this paper we present a method for 3D reconstruction of the pathological zone and propose a method for volume computing. We consider this as a start for developing a complex software tool for Deep Brain Stimulation procedure in the neurosurgery domain.

Keywords: volume/ surface rendering, multi-planar reformatting, DICOM files, segmentation

DIGITAL TECHNOLOGY IMPLEMENTATION OF PEAK CURRENT CONTROL METHOD WITH PID COMPENSATOR APPLIED TO BUCK CONVERTER

Octavian BOTEZ Mihai LUCANU Ovidiu URSARU Cristian AGHION

Abstract: Lately, digital control technology has advanced greatly in the field of power electronics[14], enabling implementation of more and more common method to control electronic converters. This paper presents the modeling and the simulation of peak current control method with PID compensator, applied to a synchronous buck converter with digital control, and its practical implementation. For simulation, the C language source code was designed for the perturbations compensating function, the duty cycle calculation applied to PWM generator, and variable current reference.

Keywords: : peak current, PID compensator, synchronous buck, digital technology

SIMULATION OF STOCHASTIC PROCESSES IN FINANCIAL MODELING

Mircea Cristian GHERMAN, Monica BORDA

Abstract: This paper discusses theoretical properties, shows the performance and presents some extensions of techniques used for simulation of stochastic differential equation applied on the financial data modeling. There are realized comparisons of different approaches for discretization schemes and their performances from the simulation convergence point of view. This study shows that, depending on the applicability of stochastic modeling to various financial data, the evolution of asset price over the time can be characterized by different processes accordingly with their dynamics.

Keywords: stochastic differential equations, simulations schemes, jump diffusion, convergence schemes

GENDER AND AGE ESTIMATION BASED ON FACIAL IMAGES

Hlaing Htake Khaung TIN

Abstract: In this paper, a fast and efficient gender and age estimation system based on facial images is developed. There are many methods have been proposed in the literature for the age estimation and gender classification. However, all of them have still disadvantage such as not complete reflection about face structure, face texture. This technique applies to both face alignment and recognition and significantly improves three aspects. First, we introduce shape description for face model. Second, the feature extraction phase, two geometric features are evaluated as the ratios of the distances between eyes, noses, and mouths. Finally, we classified the gender and age based on the association of two methods: geometric feature based method and Principal Component Analysis (PCA) method for improving the efficiency of facial feature extraction stage. The face database contains the 13 individual groups. Within a given database, all weight vectors of the persons within the same age group are averaged together. A range of an age estimation result is 15 to 70 years old, and divided into 13 classes with 5 years old range. Experimental results show that better gender classification and age estimation

Keywords: gender classification, age estimation, principal component analysis, face recognition, feature extraction

Number 2/2011

SPECTROGRAM BASED MUSICAL INSTRUMENT IDENTIFICATION USING HIDDEN MARKOV MODEL (HMM) FOR POLYPHONIC SOUND SIGNALS

Daulappa G. BHALKE, C. B. RAMA RAO, D. S. BORMANE, Manisha VIBHUTE

Abstract: Spectrogram is generated from the instrumental music, which is used to calculate the spectral, temporal & modulation features. To detect musical instruments from polyphonic music, 23 features are selected. The feature vector is generated using 12 specific features. Hidden Markov model (HMM) is used to calculate the conditional instrument existence probability. In this work, ten musical instruments from wind and string categories are used for identification. The musical instrument is recognized using HMM algorithms: Forward, Backward, Posterior decoding & Viterbi algorithm. Identification result is 91% for monophonic & 87% for polyphonic musical notes.

Keywords: Spectrogram, Feature vector, HMM, Musical Instrument Identification result

SHARING THE KNOWLEDGE: SEMANTIC MEDIAWIKI

Cornelia Flavia M. VEJA, Mircea – Florin VAIDA, Gregor HAGEDORN

Abstract: The information with the purpose of documentation, media sharing, learning, and entertainment accumulated in wikis is huge. The main problem in wikis is that the information is produced from humans to humans, not for machines. Wiki’s information is a semi-closed world and the use and re-use of wiki’s information by external non-humans agents are cumbersome tasks. Stepping further, to unstructured data but strongly connected information to structured data and annotated information is brought by adopting Semantic technologies into wikis. This paper presents a survey art and comparative study of Semantic Wikis, with the emphasis on Semantic Media Wiki model

Keywords: MediaWiki, Semantic MediaWiki, Semantic Web, Semantic wiki, Social Software, Semantic Social Software, wiki system.

REMOTE MONITORING OF THE INTELLIGENT BUILDINGS PARAMETERS

Nicolae MURESAN, Bogdan ORZA, Aurel VLAICU

Abstract: Remote monitoring of the parameters of a building requires reading/writing the values of the variables provided by networks of smart devices (sensors, actuators and controllers), that control these parameters. A solution may be the creation and optimization of a web interface that can be accessed from the Internet via a web server, providing IP connectivity and other web capabilities. Networks can also be monitored through the 220V AC network by using transceivers. Various services required in buildings can be remotely monitored, the energy management can be improved by making a more efficient use of the electricity and of the heat consumption with the aim of reducing the maintenance cost. Utilities companies (water, gas, electricity, etc.) are also becoming more interested in remote monitoring of customers' consumption and are trying to integrate some of the so called "advanced metering systems” (which contain smart meters networks). Remote monitoring of utilities, as well as monitoring building parameters, requires read/write of the network variables values of smart meters that operate with customer utility values [15], [16], [26]. Our paper presents the principles of remotely monitoring the parameters of the buildings, together with the most commonly used norms in the field and also an example of an application for remote monitoring several LonWork control networks.

Keywords: LonWorks, RNI, Router IP-852, IP Training Center, drawing LonMaker, server i.LON, X-10, PLC

HARDWARE IMPLEMENTATION OF DIVERSITY SCHEME FOR MIMO COMMUNICATION

Angela DARIE, Ion BOGDAN

Abstract: Because of the tremendous growth of wireless communication and the demand for faster communication, new applications have developed such as broadband wireless Internet access. To cope this demand for increased capacity, multiple antennas can be used for improving the performance of wireless communication systems. The aim of this paper is to implement a simple multi-antenna scheme, a 2x1 Alamouti code, on a Xilinx FPGA connected via UART to an Iris mote. Our focus is to make a practical testing for MIMO system within indoor environment using external peripherals of Iris mote and communication protocols. To accomplish this, the Iris platform has to be configured to run over I2C to the class Spartan 3E board, and the Spartan’s UART port is utilized to issue Iris radio communication stack.

Keywords: FPGA, MIMO, Iris mote, radio communication.

AN ADAPTIVE DIGITAL COMPENSATION DESIGN FOR BUCK CONVERTER TOPOLOGY

Radu ETZ, Dorin PETREUS

Abstract: This paper presents a digital type III compensator that needs no calculation made by a designer. First an analog type III compensator is designed for a buck converter and based on it the digital model is obtained. Both the analog and digital approaches are simulated in PSIM and the digital compensator is written in a Microchip dsPIC for experimental results. The dynamic response of these controllers under input voltage change and load current variations are presented.

Keywords: digital control, buck converter, type III analog compensator, type III digital compensator

ACTIVITY RECOGNITION FROM ACCELERATION DATA COLLECTED WITH A TRI-AXIAL ACCELEROMETER

Ioana FARKAS, Elena DORAN

Abstract: This paper developed a high accuracy for human activity from data collected with a single tri-axial accelerometer mounted on the right part of the hip. The accuracy of this detection is very important for classifying the postures and because of this we use acceleration magnitude and neural network method for detection and compare them to find the best solution.

Keywords: Tri-axial accelerometer, activity recognition, acceleration magnitude, neural networks

FPGA IMPLEMENTATION OF AN ACOUSTIC ECHO CANCELLER

Ioana HOMANA, Irina MURESAN, Marina TOPA, Cristian CONTAN

Abstract: Acoustic echo cancellation (AEC) is an occurrence in today communication systems. The interferences caused by acoustic echo are distracting the users and reduce the quality of the voice. In order to attenuate the effects created by the echo, a Least Mean Square algorithm is used. In this paper, we present the implementation of an LMS adaptive filter, designed on a Xilinx Virtex-5 FPGA and implemented using VHDL. ModelSIM simulation results altogether with plots obtained in Matlab prove the behavior of the chosen adaptation algorithm for the specific application.

Keywords: acoustic echo cancellation, LMS, FPGA

NONLINEAR ACOUSTIC SYSTEM IDENTIFICATION USING ADAPTIVE COMBINATIONS OF VOLTERRA FILTERS

Cristian CONTAN, Marina TOPA, Ioana HOMANA, Irina MURESAN

Abstract: The paper proposes two nonlinear system identification methods in order to reduce the compromise between convergence speed and steady-state that occurs when using third order adaptive Volterra filters. Combination of filters and combination of kernels schemes are outlined. Adaptation of the kernels is achieved using a Normalized Least Mean Square algorithm. The tradeoff between speed and stability is emphasized by selecting different step-size parameters for each kernel. Both nonlinear approaches are applied in acoustic Loudspeaker-Enclosure-Microphone setups where several sources of nonlinearities exist: the overdriven amplifier’s or small loudspeakers’ operated at high volume. A third order polynomial with distinct Linear-to-Nonlinear Ratios is chosen to model these distortions. The evaluation is made in terms of Echo Return Loss Enhancement in experiments conducted for various input signals. Results show that the overall adaptive combinations perform each time better or at least as well as the best component filter

Keywords: Nonlinear system identification, Volterra filters, adaptive combination, adaptive algorithms

A FUZZY SET GENERALIZATION OF THE EXEMPLAR-BASED IMAGE INPAINTING

Mihaela CISLARIU, Mihaela GORDAN, Aurel VLAICU

Abstract: Inpainting is a technique of modifying an image in an undetectable way, by removing large objects or damaged areas and replacing them with plausible background. In this paper we have approached the exemplar-based image inpainting method proposed by Criminisi. This method uses patches from the image to fill the damaged region. Here we propose a new fuzzy set generalization of the exemplar-based image inpainting by using a fuzzy window. We use the fuzzy window to try to improve the finding of the most similar patch for filling. Experimental results show that the proposed method yields improvements compared to other methods

Keywords: Image restoration, inpainting, fuzzy logic, RGB color space

Number 1/2011

Finger Detection in Video Sequences Using a New Sparse Representation

Vasile GUI, Gheorghe POPA, Pekka NISULA, Veijo KORHONEN

Abstract: In this paper we propose a new method for finger detection in video sequences. The method is robust and has a low computational cost. From a background subtracted image, we generate a sparse image representation, based on line strip features. We use a robust and adaptive clustering method, related to the mean shift (MS) to detect fingers and extract finger pose parameters from validated clusters. We applied the proposed method in a HCI designed to control very large public displays.

Keywords: Mean shift, sparse, robust estimation, finger detection, tracking, HCI

Programmable Control System with Applications in Direct Current Motors Control

Andrei COZMA, Dan PITICA

Abstract: This paper presents a programmable control system with applications in Direct Current (DC) motors speed control. The system consists of a signal acquisition module, a programmable high speed Digital Signal Processor (DSP), a Microcontroller Unit (MCU), an output module and a communication module. A speed and position sensor integrated in the data acquisition module provides through an I2C bus data related to the motor’s shaft speed, position and rotation direction. The DSP can perform basic arithmetic operations, finite and infinite impulse response filtering and fuzzy logic operations at high speed and with a high degree of parallelism on the data received from the signal acquisition module. The MCU is a soft processor core with a RISC instruction set and it is used for controlling the operation of all the other modules and for implementing more complicated control algorithms that cannot be performed by the DSP. The output module contains a Pulse Width Modulated (PWM) block for generating the DC motor’s control signal. The communication module transfers data to/from a PC through a USB connection. By integrating all the above mentioned modules into one single chip a complete real time control core is provided.

Keywords: automatic control, direct current motor, digital signal controller

A survey of Augmented Reality in Health Care

Marius DANCIU, Mihaela GORDAN, Aurel VLAICU, Alexandru ANTONE

Abstract: Although Augmented Reality technology was first developed over forty years ago, there are no recent studies that give an overview of the research in the field of medicine. This paper surveys the state of the art in applications of virtual environments and related technologies in the Health Care domain. Virtual Reality (VR) and Augmented Reality (AR) have the potential to assist us in fields as education, maintenance, design and reconnaissance to name only a few, but also in Health Care, improving surgical interventions and image-guided therapy by providing clinicians with interactive three-dimensional visualizations in all stages of treatment. Tools that respond to the needs of the present virtual environments systems are continuously refined and developed. Applications of these technologies improved the quality of Health Care and as we move forward into the second decade of VR, we may see more and more adopters of this technology as the cost came down. It also discuss the remaining challenges that need to be further studied and solved in order to make these systems attractive and efficient. This survey provides a starting point for anyone who work in researching or use Augmented Reality.

Keywords: Augmented reality, virtual reality, human-machine interaction, segmentation, development suite.

COMPLEX MODEL OF A SERIES LOAD INDUCTION HEATING INVERTER

T. SZELITZKY, I. INOAN, M. TULBURE, A. O. NEAGA

Abstract: The design of robust controllers require a special type of system model, called augmented plant (system) model, which includes as uncertainties the plant parameter variations. In this article accuracy of the classic describing function based model will be analyzed and updated to include the parameter uncertainties. Additive and multiplicative type parameter uncertainties will be introduced and analyzed the effect over the system model.

Keywords: Robust control, Induction heating, series load inverter

DOUBLY ITERATIVE DECODING OF SPACE-TIME TURBO CODES USING LOG-APP ALGORITHM

Ana-Mirela ROTOPANESCU, Lucian TRIFINA

Abstract: In this paper are examined the performances of a low complexity doubly iterative decoder for space-time turbo codes on a quasi-static fading channel, using a Log-APP (logarithmic a posteriori probability) compared to a Max-Log-APP (maximum Log-APP) algorithm decoding algorithm for turbo codes. The decoder uses preliminary soft values of the coded symbols, obtained after a limited number of turbo iterations, in order to reduce the spatial interference of the received signal. In the coding scheme, the core of the iterative decoding structure is a soft-input soft-output APP module. The obtained frame error rate (FER) and bit error rate (BER) performances are compared to those achieved by applying a simple scaling coefficient to the extrinsic information, when a Max-Log-APP algorithm is used.

Keywords: space-time turbo codes, doubly iterative decoder, Log-APP algorithm, Max-Log-APP algorithm, BER/FER performances

SUB-1V BANDGAP VOLTAGE REFERENCE OPTIMIZATION WITH GENETIC ALGORITHMS

Paul FARAGO, Lelia FESTILA

Abstract: This paper proposes an evolutionary approach to Bandgap Voltage Reference (BVR) design. The efficiency and feasibility of this method for solving the circuit design task are investigated. For demonstration purpose, two sub-1V BVR circuits are chosen. Circuit optimization is performed for various algorithm options. We have compared the design results obtained for purely random and deterministic initial conditions.

Keywords: Sub-1V bandgap voltage reference, evolutionary circuit design, genetic algorithm (GA), GA parameters

EFFICIENT STRATEGIES TO OPTIMIZE A POWER DISTRIBUTION NETWORK

Raul FIZESAN, Dan PITICA

Abstract: One of the biggest design challenges for today is to properly design, manufacture, simulate and validate a power distribution network (PDN) in systems with increasing speed, power dissipation and density. Power distribution networks are typically comprised of capacitors networks that have several types of capacitors and values to obtain target impedance over the required frequency range for the power/ground planes on PCBs. Capacitors provide a temporary source of localized energy for instantaneous current demands from an IC, and a low-impedance return path for high frequency noise.

Keywords: Power distribution network, target impedance, decoupling capacitor, dielectric constant, plane separation

Inertial Indoor Navigation System

Lucian Ioan IOZAN, Jussi COLLIN, Jarmo TAKALA, Corneliu RUSU

Abstract: This paper represents a first step in developing an indoor navigation system based only on Micro-Electro-Mechanical Systems (MEMS) accelerometer and gyroscope sensors. The major advantage of the proposed inertial navigation system consists on the fact that it does not require any complex infrastructure to be able to operate. For determining the distance it uses three axis accelerometer sensors, while the direction of the user is determined from the gyroscope angular rate data. As no infrastructure is used, the cost of such a system is low, but at the same time the lack of infrastructure represents its highest limitation. More exactly, for long periods of time the errors are cumulating in the navigation solution determining the deterioration of its accuracy. This limitation can be suppressed by using robust algorithms that can estimate and compensate these errors from the accelerometer and gyroscope signals.

Keywords: inertial navigation, gyroscope sensors, MEMS technology, PDR algorithm

EXPERIMENTAL MEASUREMENTS OF THE POWER CONSUMPTION FOR HARDWARE PLATFORM eZ430-RF2480

Rodica-Elena DORAN, Ioana-Iuliana FARKAS

Abstract: One of the most important characteristics that defines the fields of Wireless Sensor Networks is the focus on energy consumption. Understanding energy consumption is an essential factor for sensors networks operating on limited power reserves, since many sensors nodes are powered by batteries. Once the battery of a node has depleted the node fail making the network useless. This paper describes a series of experiments which obtained detailed measurements of the power consumption on eZ430-RF2480. The data is presented as a collection of equations for calculating the power consumption and battery lifetime in transmitting, receiving, and requesting data packets.

Keywords: sensor networks, eZ430-RF2480 platform, power consumption, periodic transmission

Number 4/2010

New pseudo-random interleaver with increased parameters

Georgian Alexandru LAZAR, Nicoleta Roxana BUZATU, Elena COJOCARIU, Lucian TRIFINA, Razvan VIERU

Abstract: This paper presents a new algorithm of obtaining a pseudo-random code-matched interleaver leading both to very good interleaver parameters and performance. The difference between the proposed permutation and other code matching techniques is that not only the distance spectrum is improved but also the parameters of the interleaver. The design procedure is described in depth, and the benchmarking is done against the High Spread-Random and the deterministic Long Term Evolution (LTE) standard interleavers.

Keywords: Interleaved coding, modulation coding, code matched, turbo codes, error correcting codes

AN EXPERIMENTAL STUDY OF QUALITY ANALYSIS METHODS IN DVB-S/S2 SYSTEMS

Alexandru-Florin ANTONE, Radu ARSINTE

Abstract: The paper is proposing a study and practical experiments of testing methods in digital television, transmitted in Digital Video Broadcasting standard. The main objective of the research presented in this paper was to establish if the testing tools designed and validated for DVB-S, are useable to evaluate DVB-S2. This is important in educational environments and small enterprises, to find new applications for the existing equipment. Testing methods follows three directions: transmitted services (data link layer), RF (Radio Frequency) interface and reception (physical layer). We pursued the following objectives: DVB-S2 model study (from Simulink Matlab), to establish the key measurement points in this evaluation, analysis of DVB-S/S2 streams, RF parameters measurements in terms of C/N (Carrier to Noise Ratio) and spectrum.

Keywords: : Digital Video Broadcasting Satellite, Test bed, Multiplex stream

SINGLE WALL NANOTUBES BEHAVIOUR IN RF FIELD

Dan PANTIS, Sorin SPINEAN, Alexandra EFIMOV, Rodica ION, Monica BORDA

Abstract: Nanomedicine represents a new discipline which combines nanoscience and nanotechnology. It opens new areas of development, like drug delivery by nanoparticles, nanotubes or fullerenes, nanostructures to be used in medical diagnostics, prevention, labeling and imaging. Our study will certificate that the temperature doesn’t have any influence above Carbon Nanotubes in our interest range (40-43º) – obtained and controlled using RF field.RF can be used in their delivery into the cell. Current radiofrequency ablation (RFA) techniques are limited by accuracy of targeting and are used for thermoablation of specific pathologies. The aim of our research was to demonstrate and to use further the fact that SWNT are ideal transporters, due to immunity at thermal variations. The biological and medical aim of the study is to evaluate this process in cancer cells. Our experiments where done into the ICECHIM Laboratory for Analysis and Testing - Laboratory testing for hazardous substances, waste and wastewater . Our results presents the mass loss correlated with degree of temperature in order to be used in the future studies as background into the cancer treatment by nanotubes delivery.

Keywords: : RF, nanotubes, cancer cell, nanobiotechnology

A MULTIMODAL AUTHENTICATION SYSTEM

Marius DAVID, Monica BORDA

Abstract: In this paper, we propose a secure fingerprint matching technique, which was combined with conventional authentication methods to achieve more secure communications. We describe in detail all the steps needed for preprocessing and fingerprint image classification. Also an algorithm based on the local structure of the minutiae is presented to match the fingerprints.

Keywords: Fingerprint, identification, authentication, normalization, filtering, binarization, skeletonization, fingerprint matching unimodal biometric system, multimodal biometric system

A CONCEPTUAL FRAMEWORK AND IMPLEMENTATION FOR DEVELOPING WS - QoS AWARE ARCHITECTURES

Cosmina IVAN, Vasile DADARLAT

Abstract: Web Services and SOA are becoming more and more popular these days and more and more businesses are planning to build their future solutions on Web Services technology. By now, SOAP and WSDL have become reliable standards in the field of Web Services execution. While the concept of UDDI allows for automatic discovery of services implementing a common public tModel interface, there have been only few attempts to find a standardized form to describe the Quality of service (QoS) with which the service is performed. Nevertheless, the QoS a provider delivers will become a decisive feature when it comes to selecting one from many availably services providing the same functionality. Today, we have sophisticated technology to actively differentiate between various QoS levels both on transport level (DiffServ, IPQoS, CoS for UMTS/ATM/RCL) and server level (load balancing, transaction differentiation, HTTP request differentiation), yet there are only few means to describe the desired QoS on application layer. To tackle this, we propose a conceptual framework based on a Xml schema to declare both clients’ QoS requirements and the QoS level service providers, and we have designed and implemented a framework for C# / .NET application developers to assign QoS requirements to a client application’s service proxy which can then be used to inquire a QoS service broker for the best QoS offer fulfilling the requirements from all offers available for services that implement the specified interface.

Keywords: Framework, QoS – Quality of Service , WS-Web Services, SOAP/WSDL/UDDI

DESIGN OF A T-SHAPED SLOT ANTENNA FOR UWB COMMUNICATIONS

Satya Kumar VANKAYALA, Saida REDDY

Abstract: This paper presents a design of T-shaped slot antenna for ultrawideband (UWB) communications. Multiple resonances are generated and unified to form a large bandwidth from 2.5 to 10.5 GHz for S11 < −10dB by constructing open ended modified T-shaped slot antenna and by extending a small section of the microstrip feed line. The simulated result shows that it has good radiation patterns and pulse handling capabilities within the band of interest.

Keywords: Ultrawideband, Microstrip feed, Group delay

NONLINEAR SYSTEM IDENTIFICATION USING ADAPTIVE VOLTERRA FILTERS FOR ECHO CANCELLING

Cristian CONTAN, Marina TOPA, Ioana HOMANA, Botond Sandor KIREI

Abstract: Adaptive nonlinear filtering plays an important role in audio signal processing and echo control. In this contribution a nonlinear system identification method is proposed. The setup is built using adaptive Volterra filters of second and third order with the same memory length. The adaptation is achieved using a Normalized Least Mean Square algorithm with a proper selected step-size parameter for convergence and stability. Functions with nonlinear characteristics are chosen to test the method. Performances are evaluated in terms of Echo Return Loss Enhancement and Mean Square Error. Results show that Volterra filters perform better than linear filters in weakly nonlinear structures.

Keywords: Adaptive algorithms, Volterra filters, nonlinear identification, error reduction

FIR SYNTHESIS OF THE HUMAN HEARING MECHANISM RESPONSE

Erwin SZOPOS, Marina TOPA, Lelia FESTILA, Horia HEDESIU

Abstract: The paper deals with the synthesis of a model for the nonlinear human hearing mechanism response. The non-uniform frequency sampling is combined with the interpolation methods to design a FIR filter to approximate the psychophysical data (audiogram). The analysis of the designed filters is achieved in LabVIEW. To evaluate the FIR synthesis the magnitude characteristic and the impulse response for different interpolation methods are compared.

Keywords: Audiogram, psychophysical, approximate FIR filter, GCD, interpolation, relative error

32x32 Parallel Analog Architecture For Image Processing Using Log Domain Active Pixel

Ion VORNICU, Liviu GORAS

Abstract: In this paper the transistor level simulation of a 32x32 analog parallel architecture of Cellular Neural Network (CNN) type able to perform spatial filtering operations is reported. The circuit level implementation is obtained by the log-domain mapping of the linear equations describing the CNN by using MOS transistors biased in weak inversion. The spatial filter is simulated in a 340nm CMOS technology. In the last part of the paper several examples of image processing operations like smoothing and edge detection using a chessboard type input image are presented.

Keywords: spatial filtering, log-domain, image processing, smoothing, edge detection

GENETIC ALGORITHMS FOR NETWORK CODING OPTIMIZATION

Zsuzsanna KISS, Zsolt POLGAR, Mircea GIURGIU

Abstract: The optimization of networks which implement Network Coding techniques is necessary for efficient resource utilization. If the data is encoded only in a subset of nodes, selected in the appropriate way, the complexity of the coding operations could be decreased, while ensuring the imposed performances. Finding the minimum number of encoding nodes is a NP-hard problem, but genetic algorithms could offer a good solution to this problem. In this paper we investigate by computer simulations the performances of a genetic algorithm adapted to the network optimization problem and we try to find the combination of parameters which ensures low optimization time.

Keywords: Network Coding, genetic algorithms, network optimization, node selection

COMMUNICATION BETWEEN INTELLIGENT DEVICES IN LONWORKS CONTROL NETWORKS

Nicolae MURESAN, Bogdan ORZA, Aurel VLAICU

Abstract. A modern control network consists of a number of intelligent devices that communicate on one or more channels, using a common protocol. The network’s intelligent devices, called nodes, may each include more microprocessors running the application, and the protocol. Also, each device includes a component called transceiver, that provide the electrical interface with the communication channel. In the LonWorks networks, most of the devices (intelligent sensors, controllers, actuators), exploit the advantages of the Neuron Chip, the default LonWorks node, which includes three processors, two of which execute the LonTalk protocol, and the third the device’s application-programme. The Neuron Chip supplies the functions of the first six OSI stack layers, the user needing to provide only the configuration and the programming of the application layer. The goal of this paper is to present the comunication mode between intelligent devices, in LonWorks networks, at the MAC and Application levels, based on LonTalk protocol. This analysis is useful in order to know the conditions to be fulfilled for an optimal design of the network, optimal setup and configuration of the devices, and to provide multivendor interoperability. The paper also presents an example of effective design of a lighting control network.

Keywords: LonWorks, LonTalk, LonMaker Basic Shape, LonMaker design, predictive p-persistent CSMA.

Number 3/2010

Speech and Signature Based Multi-Modal Biometrics Verification

Mohamed SOLTANE

Abstract—This paper describes a combined behavioral techniques based on speech and signature biometrics modalities. Fusion of multiple biometric modalities for human verification performance improvement has received considerable attention. Multi-biometric systems, which consolidate information from multiple biometric sources, are gaining popularity because they are able to overcome limitations such as non-universality, noisy sensor data, large intra-user variations and susceptibility to spoof attacks that are commonly encountered in mono modal biometric systems. Soft decision level fusion based Gaussian mixture models (GMM), in which the (EM), (GEM) and (FJ) algorithms for estimating the parameters of the mixture model and the number of mixture components have been compared. The test performance of the fusion, EER=0.0 % for "EM" and "FJ", EER=0.02 % for "GEM", show that the combined behavioral information scheme is more robust and have a discriminating power, which can be explored for identity authentication.

Keywords: Biometric authentication, behavioral biometrics, speech analysis, signature verification, soft decision fusion, Gaussian Mixture Modal, EM, GEM and FJ.

BANDGAP VOLTAGE REFERENCES BASED ON F-F-1 FUNCTIONS

Lelia FESTILA, Mihaela CIRLUGEA, Lorand SZOLGA

Abstract: We propose in this paper a bandgap voltage reference (BVR) structure based on a class of circuits belonging to ELIN design [5]. F-F^-1 amplifiers[6] were used for this purpose. They have the advantage of simplicity, a natural nonlinearity compensation and offer possibility to be current controlled or tuned so they do not need resistors as usually bandgap references do. One of the two BVR schematics resulted from the general structure is a simplified variant of a reconfigurable BVR proposed in literature [4]. The second resulted and proposed variant is a new one. Simulations proved the functionality and validity of this general structure and BVR unconventional variants.

Keywords: Bandgap voltage references, ELIN design, F-F^-1 functions

A Review of Current Strained Silicon nanoscale MOSFET Structures

Amit CHAUDHRY, Garima JOSHI, J. N. ROY, D. N. SINGH

Abstract: Improvement in performance of Si- Metal Oxide Semiconductor Field Effect Transistors (MOSFET) through scaling is very difficult due to the several physical phenomenon mainly non classical in nature at nanometer scales. In this paper, an attempt has been made to give a detailed review of strained silicon technology so far developed since its inception. Various biaxial and uniaxial strained (MOSFET) structures have been studied. The advantages, disadvantages, physics and fabrication process of these structures have also been outlined and reviewed critically. The review indicates that this technology is very much required in nanoscale MOSFETs due to its several potential benefits over the conventional bulk technology. The paper also provides the insight of how the use of various techniques such as relaxed SiGe, graded SiGe, strained silicon on insulator (SSOI), silicon germanium on insulator (SGOI) and various other hetero-structures help in providing improved performance.

Keywords: Uniaxial strain, Biaxial strain, Mobility, SiGe, strained silicon technology, strained silicon on insulator, Hetero-structures.

USING SOFTWARE DEFINED RADIO AND COGNITIVE RADIO TECHNOLOGIES IN SMART HOME ENVIRONMENTS

Levente SZALONTAI, Lelia FESTILA

Abstract: Nowadays there’s a big need for intelligent smart home environments to reach a cost effective energy management and in the same time to provide a high level of comfort and security for smart home inhabitants. In smart home environments there are many different wireless sensor, home appliance and multimedia networks working on different airinterfaces, protocols and standards that have a contribution to this cost effective energy management. The biggest problem is the lack of flexibility regarding the communication between these networks and smart home network devices control. The solution which seems to be the best one is using Software Defined Radio (SDR) and Cognitive Radio (CR) technologies which could provide higher flexibility in controlling and monitoring these networks as well as much faster implementation of new standards and protocols. We analyze in this paper the possibilities for using these two technologies in modern smart home wireless networks. Starting from predefined tasks a modern smart home should fulfill, we made a critical analysis of SDR technology’s current state of the art and finalized this analysis with conclusions regarding the possibility of using one of the best solutions to implement this technology in smart home networks.

Keywords: software defined radio, smart home, cognitive radio

On Hand Gestures Recognition Using Hidden Markov Models

Vasilica TATARU, Laurentiu Radu VIERIU, Liviu GORAS

Abstract: : In this paper several results concerning static hand gesture recognition using an algorithm based on left-right Hidden Markov Models (HMM) are presented. The features used as observables in the training as well as in the recognition phases are based either on the 2D Discrete Cosine Transform (DCT) or on the Principal Component Analysis (PCA). The left-right topology of the HMM together with the Baum-Welch algorithm for training and Viterbi algorithm for testing led to the best results. Simulation results show that the system has a recognition rate of 97.5% for DCT and 95% for PCA.

Keywords: Human-Computer Interaction, Hand Gestures Recognition, Hidden Markov Models

ITERATIVE REWARDS GENERATOR FOR MOBILE TELECOMMUNICATIONS OPERATORS

Marius Cristian URECHE, Mircea VAIDA, Alin VOINA

Abstract: Dealing with large amounts of data always was a delicate operation. Mobile telecommunications operators need high quality management systems to provide top-class services for their customers. The challenge is to assure data processing in a real-time or nearly real-time manner while facing the dynamic changes that are the heart of this field of engineering. We propose one targeted software architecture for iterative rewards generator enabling mobile telecommunications operators to grow their revenue by implementing iterative rewarding campaigns.

Keywords: mobile telecommunications, rewards generator, intelligent network platforms, reward-based loyalty campaigns

LABVIEW FPGA IMPLEMENTATION OF DIGITAL REVERBERATORS

Ioan CATUNA, Erwin SZOPOS, Marina TOPA

Abstract: : The goal of this paper is to present a way in which digital reverberation algorithms can be implemented on FPGA. Schroeder’s late reverberator was chosen to be analyzed, implemented as an online algorithm in LabVIEW FPGA and then deployed on a National Instruments CompactRIO system. The hardware implementation is tested with three different voice sequences, and its parameters are adjusted so that the reverberation effect improves the quality of the sound, without inserting distortions or unwanted noise.

Keywords: : Comb and all-pass filters, Schroeder’s reverberator, loop and reverberation times, LabVIEW FPGA

An Approach on Bimodal Biometric Systems

Eugen LUPU, Simina EMERICH

Abstract: : Although unimodal biometric systems have an edge over traditional security methods they have their limits and raise a variety

of problems. In order to overcome some inconveniencies and limitations of these systems and in order to realize reliable and robust authentication systems, the use of multimodal biometric systems is recommended. The information contained in multiple biometrics can be integrated by using various methods at distinct levels (rank, decision, match-score levels, feature and sensor) and in different contexts. A short presentation of the different fusion techniques has been made to highlight their advantages and disadvantages. The experiments were made on a bimodal biometric system employing speech and dynamic signature as biometric features. The results show that the score fusion of the two biometric improves the system overall performance.

Keywords: : biometric, bimodal, speech, dynamic-signature, score level fusion, normalization

A THREE LEVEL VOLTAGE INVERTER USED FOR AN INDUCTION MOTOR CONTROL

Nicoleta Roxana BUZATU, Dimitrie ALEXA, Georgian Alexandru LAZAR

Abstract: : Advances in power electronics technology allowed the wide investigation of multilevel converters that provide higher power with less harmonic components compared to the two - level structure. Among this family of converters, the three - level neutral – point - clamped (NPC) converter is widely used in medium voltage applications. This paper presents a 3 - level Neutral – Point - Clamped inverter using space vector pulse width modulation (SVPWM) for an induction motor control.

Keywords: : Three - level voltage inverter, neutral-point-clamped (NPC), induction motor, speed control

PERFORMANCE ANALYSIS OF ASYMMETRIC TURBO CODED MODULATION WITH TRANSMIT ANTENNA DIVERSITY

Lucian TRIFINA, Daniela TARNICERIU, Ana-Mirela ROTOPANESCU

Abstract: The performances of turbo coded modulation using asymmetric turbo codes with transmit antenna diversity are analyzed. The component convolutional codes are considered to have memory 2 and 3, respectively, and their generator polynomials are both primitive and nonprimitive. The performances of asymmetric turbo coded modulation were evaluated performing simulations to obtain the bit error rate (BER) and the frame error rate (FER). Primitive polynomials lead to better performances for FER, whereas the nonprimitive ones lead to slightly improvements in low SNR range.

Keywords: turbo coded modulation, antenna diversity, asymmetric turbo codes

CAPACITY ANALYSES FOR IEEE 802.11N NARROWBAND AND WIDEBAND CHANNEL MODELING

Irina I VERMESAN, Tudor PALADE, Rebeca M COLDA

Abstract: : In this paper we investigate the channel capacity when MIMO (Multiple Input Multiple Output) technology is employed for wireless communication systems that are designed under IEEE 802.11n conditions. First, we emphasis the methods of modeling the NLOS (Non Line of Sight) narrowband and wideband communication channel. Afterwards, the channel capacity behavior is analyzed as a function of ray propagation physical parameters given in the standard, such as: AoA/AoD (Angle of Arrival/ Angle of Departure), AS (Angle Spread), number of clusters, number of antennas and inter – element distance. The spatial correlation properties are also incorporated in our investigation.

Keywords: : narrowband channel modeling, wideband, capacity, spatial correlation

FILTER BANK SYNTHESIS FOR ADAPTED WAVELET ANALYSIS

Zoltan German SALLO

Abstract: : The multiresolution theory of orthogonal wavelets proves that any conjugate mirror filter characterizes a wavelet that generates an orthonormal basis of L2R. This paper presents a method to obtain a discrete wavelet function, through FIR filter synthesis starting from a random discrete sequence. This sequence is used as a FIR filter which generates the four conjugate mirror filters to implement the discrete wavelet transform. The refinement of this sequence is carried out by a learning structure which follows to minimize the error between the signal and its reconstructed form from first order decomposition.

Keywords: : wavelet analysis, filter banks, perfect reconstruction

DEVELOPMENT FRAMEWORK FOR CONGESTION AVOIDANCE MECHANISMS

Vasile Teodor DADARLAT, Raluca JELER, Adrian PECULEA, Bogdan IANCU, Emil CEBUC, Cosmin ARDELEAN

Abstract: The paper is presenting a novel approach for the design and implementation of a development framework for congestion avoidance mechanisms. A method for dynamically adjustment of the queues length function of their average queue size is also proposed. The developed algorithm, called improved WRED, uses this method which allows for a better use of the bandwidth of the link. Both the proposed algorithm and the traditional WRED were tested using the framework for the development of congestion avoidance mechanisms. The experiments showed that the improved WRED has better performance than the traditional.

Keywords: framework, WRED, QoS, congestion avoidance, algorithm.

MODIFIED DIJKSTRA'S ALGORITHM WITH CROSS-LAYER QOS

Andrei Bogdan RUS, Virgil DOBROTA, Adrian VEDINAS, Georgeta BOANEA, Melinda BARABAS

Abstract: This paper presents a modified Dijkstra’s algorithm that calculates the distance between multiple sources and a single destination. It corrects the deficiencies of the classical approach by taking into account the dynamicity of the QoS parameters at the Physical Layer and MAC Sub-layer. The proposed composite metric is based on the available transfer rate, one-way delay and bit error rate, all of them measured or calculated in real time due to a Cross-Layer QoS software module. The proof-of- concept was obtained by simulations in OMNET++.

Keywords: Cross-layer, modified Dijkstra, QoS parameters

Number 2/2010

ACTIVE BALANCING METHOD FOR BATTERY CELL EQUALIZATION

Dorin CADAR, Dorin PETREUŞ, Toma PĂTĂRĂU, Niculaie PALAGHIŢĂ

Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology,

26-28 G. Bariţiu Street, 400027 Cluj-Napoca, Romania, Phone +40264401415, E-mail: Dorin.Cadar@ael.utcluj.ro

Abstract: The emerging need for more power has led to the development of series-parallel connected battery cells. Having a more complex system also brought with it new challenges. One of these is the problem of cell imbalance. Balancing algorithms are based on extracting or adding charge in order to have balanced cells. Their goal is to protect the battery from damage and to help prolong the battery’s life. Without a balancing system the individual cell voltage will drift away, the total capacity of the pack will decrease more rapidly during operation and also the estimation of the state of charge will be unreal. This paper presents the theory behind the most important balancing algorithms developed so far and also proposes another method of cell balancing, method tested both through simulation and implementation.

Keywords: battery cell, balancing algorithm, charge

BATTERY AND SUPERCAPACITOR CHARGER BASED ON LLC CONVERTER

Dorin PETREUŞ, Toma PĂTĂRĂU, Ionuţ CIOCAN

Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology,

26-28 G. Bariţiu Street, 400027 Cluj-Napoca, Romania, Phone +40264401415, E-mail: Dorin.Cadar@ael.utcluj.ro

Abstract: The need for a small economical power supply for battery chargers has increased in the last couple of years. A LLC converter is used as a battery charger due to its low EMI noise, high power integration and high efficiency. The paper presents a more accurate approach for the design and analysis of the LLC converter. A comparison between the calculations, the simulations and the practical results is also presented. A new analysis and design method which enables the use of the LLC converter as a battery charger and super capacitor charger is presented.

Keywords: resonant converters, battery charger, ZVT, ZCT.

Acoustic Analysis of A Room

Norbert TOMA, Marina Dana ŢOPA, Botond Sandor KIREI, Ioana HOMĂNĂ

Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca

Str. G. Bariţiu nr. 26-28, 400027 Cluj-Napoca, Romania, Tel: +40264401243, Fax: +40264591689, Marina.Topa@bel.utcluj.ro

Abstract: The paper presents the acoustic analysis of a room using both software and hardware tools. First several methods for acoustic parameters estimation were tested. The most accurate measurements were used for building 3D images of the room’s impulse response. Next the room was analyzed in realistic conditions and maps of acoustic parameters were plotted. Conclusions regarding the acoustics of the room were drawn.

Keywords: room acoustics, impulse response, acoustic parameters, plenacoustic function, acoustic sources

A SIMULATION OF THE PERFECTLY MATCHED LAYER IN THE 3-D CASE

Ioana SĂRĂCUŢ, Victor POPESCU, Dan Ovidiu MICU

Technical University of Cluj-Napoca, G.Baritiu Street 26-28, Cluj-Napoca, phone 0264-401803

E-mails: Ioana.Saracut@bel.utcluj.ro , Victor.Popescu@bel.utcluj.ro , DanO.Micu@et.utcluj.ro

Abstract: The Perfectly Matched Layer (PML) technique, introduced by J. P. Berenger in 1994, is used for solving unbounded space problems. In this paper we implemented in Matlab a three dimensional Finite-Difference Time-Domain coordinate space surrounded by PML. First we evaluated the performances of the PML when a plane wave having a Gaussian pulse profile propagates. Then a dipole antenna is simulated in the same computational space and some simulation results, including the radiation pattern, are presented.

Keywords: Finite-Difference Time-Domain, Perfectly Matched Layer, electromagnetic field, dipole antenna

WORKFLOW BASED SERVICE COMPOSITION USING GENETIC ALGORITHMS

Florin-Claudiu POP, Marcel CREMENE, Mircea-Florin VAIDA

The Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology

Department of Communications, 26 – 28 Baritiu, 400027 Cluj-Napoca, Romania, Phone: (+40) 264 401810

Email: { Florin.Pop, Cremene, Mircea.Vaida }@com.utcluj.ro

Abstract: One common direction in service oriented computing is to use predefined workflows to compose new services that combine the functionality of existing services. But the problem with this approach is that services can’t be composed for which workflows are not available. We propose a genetic algorithm that combines workflows in order to find the best solution that satisfies a set of objectives formulated using imperative natural language requests. The function that measures how close a solution is to what the user expects from the composed service is a semantic similarity function, called the conceptual distance.

Keywords: web services, service composition, semantic similarity, genetic algorithms

A NOVEL END-TO-END QOS FRAMEWORK OVER HETEROGENEOUS NETWORKS - AN ARCHITECTURAL APPROACH

Bogdan IANCU, Vasile DADARLAT, Adrian PECULEA

Technical University of Cluj-Napoca, Faculty of Automation and Computer Science, Computer Science Department

26-28 G. Baritiu street, Cluj-Napoca, Romania, Tel: +40264202369, {Bogdan.Iancu, Vasile.Dadarlat, Adrian.Peculea}@cs.utcluj.ro

Abstract: End-to-end QoS (Quality of Service) provisioning over heterogeneous networks is a major challenge. The poor communication between Service Providers, the increasing complexity of the supported services, the need for fulfillment of end-to-end requirements represent the main problems to be solved by the new QoS frameworks. This paper presents a novel system architecture for an end-to-end QoS framework over heterogeneous networks, that aims to guarantee network performances and to provision end-to-end QoS over multiple autonomous systems.

Keywords: end-to-end QoS, heterogeneous networks, architecture, autonomous systems.

TLS PROTOCOL: SECURE PROTOCOL WITH CLIENT PUZZLES

Raluca CATARGIU, Monica BORDA

Faculty of Electronics,Telecommunications and Information Technology , Cluj-Napoca

Technical University of Cluj-Napoca, 26-28 Bariţiu Street, 400027, ralu_catargiu@yahoo.com

Abstract: Client puzzles are commonly proposed as a solution to denial-of-service attacks. We analyzed the TLS protocol and implement Client Puzzles algorithm. We pursued the following objectives: analyze TLS handshake protocol and implement Client Puzzles algorithm using Hash Functions, Diffie-Hellman Base Puzzle Scheme and Threshold Puzzles. The originality of the paper is due to a comparative study of methods used to create the Client Puzzle solution. We also present measurements of execution time when our modified protocol is under attack.

Keywords: client puzzles, attack, threshold client puzzles.

USING DYNAMIC TIME WARPING ALGORITHM OPTIMIZATION FOR FAST HUMAN ACTION RECOGNITION

Tamás VAJDA

Sapientia Hungarian University of Transylvania, Tîrgu Mures, Romania

Abstract: In this paper, we present an approach based on dynamic programming and neural network for recognition and matching human action. Each human action is represented by the angular motion of the body parts. Each body part angular motion is represented by an one-dimensional time series. These time series are then compared separately for every body part with templates using dynamic programming (DTW). The results of the comparisons are used as input for a neural network that classifies the human action.

Keywords: Dynamic Time Warping, action recognition.

ANALYSING THE STABILITY OF CIRCUITS BASED ON OPERATIONAL AMPLIFIERS BY USING FREQUENCY-DOMAIN SIMULATIONS

Marius NEAG, Raul ONEŢ, Marina ŢOPA

Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology

26, G. Baritiu St, Cluj-Napoca, Romania, Phone: +40264 401470, Fax:+40264 591689, Marius.Neag@bel.utcluj.ro

Abstract: This paper analyzes in detail two of the most popular methods of determining the loop gain of OpAmp-based feedback circuits through frequency-domain Spice simulations. The limitations of the simpler method – that involves breaking the feedback loop by inserting an independent voltage source with DC=0 and AC=1 – are highlighted in comparison with a more precise method, based on the Rosenstark theorem. The discussion encompasses all types of amplifiers: the traditional (V-V) OpAmp, the Current-Feedback OpAmp (CFB-OA), the transconductance OpAmp (OTA) and the Current – Current OpAmp with asymmetric inputs (such as a second-generation current conveyor). Recommendations are made based on analytical analysis and simulation results.

Keywords: OpAmp based feedback circuits, stability, small-signal loop gain, module and phase margin, Spice simulations

Number 1/2010

POLYMORPHIC AND METAMORPHIC CODE APPLICATIONS IN PORTABLE EXECUTABLE FILES PROTECTION

Liviu PETREAN

“Emil Racoviţă” High School Baia Mare, 56 V. Alecsandri

Abstract: Assembly code obfuscation is one of the most popular ways used by software developers to protect their intellectual property. This paper is reviewing the methods of software security employing metamorphic and polymorphic code transformations used mostly by computer viruses.

Key words: code, polymorphic, portable

DESIGN OF A PROGRAMMABLE CONTROL SYSTEM

Andrei COZMA Dan PITICA

Applied Electronics Department, Technical University of Cluj Napoca, Romania

E-mail: andrei.cozma@ael.utcluj.ro, dan.pitica@ael.utcluj.ro

Abstract: This paper presents the design of a system intended to be used for automatic control applications. The system consists of a signal acquisition module, a programmable high speed Digital Signal Processing (DSP) core, a Microcontroller Unit (MCU), an output module and a communication module. Both analog and digital signals can be given as input to the system. The analog signals are digitized by a high speed analog to digital converter, while the digital signals are read through an I2C bus. The programmable DSP core can perform basic arithmetic operations, finite and infinite impulse response filtering and fuzzy logic operations at high speed and with a high degree of parallelism on the data received from the signal acquisition module. The MCU is a soft processor core with a RISC instruction set and it is used for controlling the operation of all the other modules and for implementing more complicated control algorithms that cannot be performed by the DSP core. The output module contains multiple Pulse Width Modulated (PWM) signal generators and high speed Digital to Analog Converters (DAC) for analog control signals generation from the digital control signals received from the MCU or DSP core. The communication module transfers data to/from a PC through a USB connection. By integrating all the above mentioned modules into one single chip a complete real time control core is provided.

Keywords: automatic control, microcontroller, digital signal processor

TEXT-INDEPENDENT SPEAKER VERIFICATION: A COMPARATIVE ANALYSIS STUDY

Mohamed SOLTANE, Noureddine DOGHMANE and Noureddine GUERSI

Electronics Department, Faculty of engineering science

Badji Mokhtar University of Annaba, 23000 Annaba - Algeria

soltane@lri-annaba.net & xor99@hotmail.com , ndoghmane@univ-annaba.org, guersi54@yahoo.fr

Abstract: Gaussian mixture models (GMMs) remain the state of the art technique for modeling spectral envelope features for speech recognition systems. This paper presents a comparative analysis of the performance of three estimation algorithms Expectation Maximization (EM), Greedy EM Algorithm (GEM) and Figueiredo-Jain Algorithm (FJ) based Gaussian mixture models (GMMs) for text-independent speech biometrics verification. The simulation results are showed significant performance achievements. The test performance of, EER=0.26 % for "EM", EER=0.21 % for "GEM" and EER=0.16 % for "FJ", show that the behavioral information scheme of speech biometrics is more robust and have a discriminating power, which can be explored for identity authentication.

Keywords: Biometric authentication, behavioral, speech, soft decision and Gaussian Mixture Modal, EM, GEM and FJ.

ASYMMETRIC DIFFUSION AND FUSION TECHNIQUES FOR IMAGE RESTORATION

Romulus TEREBES¹ Monica BORDA¹ Sorin POP¹ Cosmin LUDUSAN^1,2 Olivier LAVIALLE²

¹Technical University of Cluj-Napoca, ²CNRS, IMS, UMR 5218

¹15,C.Daicoviciu Street, +40-264-401564, ²351 Cours de la libération, 33405 Talence, France

Romulus.Terebes@com.utcluj.ro

Abstract: We propose a novel method for restoring and enhancing images. The method is based on the Partial Differential Equations (PDE) framework and image fusion techniques. The degraded image is processed independently with two PDE based diffusion filters with different theoretical properties, an asymmetric forward diffusion equation and an evolved shock filter, and then fusion is performed at an intermediate results level. The procedure is iterated for a predefined number of steps and the restored or enhanced image is obtained as the output of the last fusion step. Through application samples we show that the method can be successfully applied for restoring images composed of oriented patterns and degraded by additive Gaussian noise and Gaussian blur.

Keywords: Diffusion, partial differential equations, shock filters, image fusion

DNA PATTERNS LOCALISATION USING SPECTROGRAMS

Alin VOINA, Petre G. POP

Comm. Dept., Technical University of Cluj-Napoca

26-28 Baritiu str. 400027, Cluj-Napoca, ROMANIA, e-mail: alin.voina@com.utcluj.ro

Abstract. This paper presents the possibilities for processing and analysis of genomic sequences which are converted into a numerical representation. There were used two types of symbol mapping: indicator sequences and quartic mapping. The resulted sequences were analyzed using a spectral analysis, completed with the obtaining of BW spectrogram, technique which offers the possibility for a unique and innovative visualization of periodical properties of genomic sequence. The advantages of this kind of representation are that offers an overview of the whole genomic sequence without using any a priori data. Once we have the spectrogram, we can obtain approximate information about positions and length of the repeated sequences.

Keywords: DNA repeats,spectral analysis, Fourier transform, spectrogram, genomic sequence

A STRUCTURE FOR HARD HANDOVER ANALYSIS USING AGENTS TECHNOLOGY

Gabriel SIRBU¹, Ion BOGDAN²

¹ Dunarea de Jos University of Galati, Electronics and Telecommunications Dept., Domneasca-47 Str., Galati-800008, Romania, email gabriel.sirbu@ugal.ro

² Gheorghe Asachi Technical University of Iasi, Telecommunications Dept., Blvd. Carol I - 11, Iasi-700506, Romania, email bogdani@etc.tuiasi.ro

Abstract: The mobility of the users in a cellular mobile environment is involving the use of the radio resources. The main aspect, which characterizes this attribute, is the hard handover and this must be carried out without blocking the existing calls of the static positioned users. Combining the two sets of users, namely static positioned users and mobile users, the resulting system must be able to manage the conflict situations, resulting in a minimum blocking probability, achieved with a centralized structure, or a distributed structure. Our work is introducing a minimal distributed structure that can be used to analyze the behavior of a cellular mobile system, based on the agents technology. The structure includes two agents and each of them implements the entire operations specific to a dynamic channel allocation scheme, into a cell. The third agent is added in order to manage the eventual hard handovers requests. This three agent minimal structure can be then multiplied in order to extend to, to simulate and to analyze various real conditions.

Keywords: cellular communications, hard handover, user mobility, channel management, Dynamic Channel Allocation Scheme, cost function, MASE technology, fixed agents, blocked call probability, forced call probability.

Number 4/2009

MEDICAL DIAGNOSIS SYSTEM BASED ON IRIS ANALYSIS

Adrian LODIN¹ Sorina DEMEA²

¹Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology,

26-28 G. Baritiu Street, Tel. +40 264 401804, Fax. +40 264 591340, E-mail: Adrian.Lodin@bel.utcluj.ro

²Oftalmological Center of Investigation REVIEW, 9 Motilor Street, Tel/fax +40 364 107206 Cluj-Napoca, Romania

Abstract: The goal of this study is to present a system which is used to process iris images for medical purposes. Such a system should assist a diagnosis based on correlation between medical pathology and different sectors from the surface of the iris. Iris surface texture regions, which offer useful information, can be revealed using computerized texture and color analysis. Iridologists work showed that locations of those regions upon segmented iridology charts point out an interrelation between map sectors and internal body system medical pathology. Our goal is to automatically identify those regions and established an assist diagnosis. The final automatically generated diagnosis needs user approval, thus making the system semi-automatic.

Key words: iris recognition, medical pathology, texture regions, diagnosis, Hough Transform.

LABVIEW FPGA BASED NOISE CANCELLING USING THE LMS ADAPTIVE ALGORITHM

Erwin SZOPOS¹ Horia HEDESIU²

¹Bases of Electronics Department, ²Electrical Machines, Marketing and Management Department,

Technical University of Cluj-Napoca, Romania

26-28 G. Baritiu str., Cluj-Napoca, Romania, Tel: +40264401803; Fax: +40264591340 Erwin.Szopos@bel.utcluj.ro, Horia.Hedesiu@mae.utcluj.ro

Abstract: This paper proposes an architecture for implementing the Least Mean Square (LMS) adaptive algorithm, using a 20 bit fixed-point arithmetic representation. The architecture length was established to 16, but it can be easily modified. This is an advantage for large filter orders. The method can also be applied to other LMS versions. This architecture is implemented using the NI cRIO-9104 FPGA chassis. The NI cRIO-9012 is a real-time module used for signal control and storage. Experiments were done regarding signal to noise ratio (SNR), filter length and type of input signals. The obtained results indicate the implemented algorithm as having high performance, while still incurring some limitations. The design is evaluated in terms of SNR, filter length and FPGA resources.

Keywords: LMS, FPGA, SNR, real-time module, noise cancelling, filter length, step-size.

A NOVEL APPROACH FOR EFFICIENCY ENHANCEMENT AND SIZE MINIATURIZATION OF UWB PRINTED ANTENNAS

Ramu PILLALAMARI ¹⁾, Sasibhushana Rao GOTTAPU² S. Srinivasa KUMAR¹

¹⁾Reliance Technology and Innovation Centre, Reliance Communications Ltd. India

¹⁾Radio Frequency and Microwave Laboratory, Dept. of Electronics and Communication Engineering, Jawaharlal Nehru Technological University Kakinada, Andhra Pradesh, India

²Andhra University, Visakhapatnam, Andhra Pradesh, India, Corresponding Author: ramu06410230@gmail.com

Abstract: The printed antenna is one of the most preferred antenna structures for low cost and compact design of wireless communication systems. In this paper we have investigated a new approach for improving the radiation efficiency and performance of the antennas with the size miniaturization, in particular we have simulated two types of UWB printed monopole antennas for proving this approach with results: circular patch and Compact Miniaturized Semi-Circular patch UWB monopole antennas and in detail investigation was presented on size miniaturization in order to achieve very compactness and high radiation efficiency with out degradation of the functional parameters and overall performance. Simple rectangular microstrip lines are used for feeding the printed monopole antennas. This UWB monopole antenna designed works well for the whole UWB frequency band 3.1-10.6GHz.

Key words: UWB, Semicircular, Miniaturization printed monopole antennas, Microstrip lines.

FORECASTING BY NEURAL NETWORKS IN THE WAVELET DOMAIN

Ion RĂILEAN, Sorin MOGA, Monica BORDA

Technical University of Cluj-Napoca, 15 Constantin Daicoviciu Street,Cluj Napoca,400020,Romania

Telecom Bretagne, Technopôle Brest-Iroise - CS 83818 - 29238 Brest Cedex 3 – France

Email : raileanion@yahoo.com

Abstract: This paper presents a forecasting method for time series. This method combines the wavelet analysis and several forecasting techniques such as Artificial Neural Networks (ANN), linear regression and random walk. The proposed method is tested using three real time series: the first contains historical data recorded during eight weeks from a WiMAX network and the other two are based on financial series. It is shown that AI with wavelet analysis can be an efficient and versatile approach in time series prediction for small periods time interval (up to 1 month). For long time interval, the best method used is Linear Regression technique. Also we compared the results obtained using various types of wavelets. The results show that Daubechies 1 (db1) and Reverse biorthogonal 1 (rbio1.1) give the best results.

Key words: Time series, Wavelet transform, forecasting, Neural Networks

A PSPICE MODEL FOR T MATCHING CIRCUITS

Ioana SĂRĂCUŢ Victor POPESCU

Technical University of Cluj-Napoca,

G.Baritiu Street 26-28, Cluj-Napoca, phone 0264-401803

E-mail: Ioana.Saracut@bel.utcluj.ro, Victor.Popescu@bel.utcluj.roIoana.Saracut@bel.utcluj.roVictor.Popescu@bel.utcluj.ro

Abstract: In electronics, impedance matching is an engineering technique employed in circuit design for matching unequal source and load impedances. This can be done by inserting a matching circuit between the source and a load. For given values of the source resistance, load resistance, the modulus of the coupling factor and the frequency at which the matching is achieved, four different circuit result. In this paper, we simplified the calculation of the reactances of the elements and presented a PSpice model that can be used to model any of the four matching circuits. Hence the model can be used as an impedance matching circuit without the need to calculate the nature and the values of its elements. Then we compared the time and frequency behaviors of a circuit that uses this PSpice model to the case when the same circuit doesn’t use it.

Key words: propagation, matching circuit, working frequency, reflected wave.

EVALUATING THE PERFORMANCES OF THE CASTGATE TUNNEL SERVER OVER TCP AND UDP LINKS IN MULTI-CLIENT CONFIGURATION

Melinda BARABAS Georgeta L. BOANEA Kris STEENHAUT* Virgil DOBROTA

Technical University of Cluj-Napoca, Communications Department

26-28 George Baritiu Street, 400027 Cluj-Napoca, Romania, Tel: +40-264-401226

E-mails: {Melinda.Barabas, Georgeta.Boanea, Virgil.Dobrota}@com.utcluj.ro

* Vrije Universiteit Brussel, ETRO Department, Pleinlaan 2, B-1050 Brussels, Belgium

E-mail: Kris.Steenhaut@vub.ac.be

Abstract: The CastGate architecture defines a transitory tunneling solution, offering access to native multicast. This paper describes the implementation of a testing application (CastGate Tester - CT) for the CastGate Tunnel Server (TS), using TCP and UDP links. The goal is to evaluate the performance of the TS in a controlled environment, by emulating a variable number of clients that connect to the server. Two scenarios were defined: load test and stress test. Based on these, the limitations of the CastGate Tunnel Server were identified. The obtained results can be used in the optimization stages of the TS.

Key words: multicast, CastGate project, UMTP*, Tunnel Server, Tunnel Client, load test, stress test.

ASPECTS OF ALGORITHM DEVELOPMENT FOR SYSTEMS WITH ONLINE DETECTION AND RECOGNITION OF TV COMMERCIALS

Radu ARSINTE

Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology, Communications Department, Tel: +40-264-595699, Str. Baritiu 26-28, Radu.Arsinte@com.utcluj.ro

Abstract: The paper describes a series of research stages in the development process for an automated system, able to detect commercial brakes in TV signals, and recognize a basic set of logos form the TV commercials. First, the conceptual diagram of the system is presented. Theoretical aspects of the recognition process are presented, and there are based on image matching algorithms. The paper presents also the test bench implemented to verify the correctness of the algorithm. The test bench is based on EasyAccess environment from eVision. Using the described test bench it is possible to tune the main parameters of the algorithm and realize extensive tests for different conditions and images. Using this test bench made possible to realize over 600 matching experiments using 20 logos of the most important international brands and 30 images captured in real conditions using different TV tuners.

Key words: Video acquisition, Commercial brake, Logo Detection, Video Streaming

COMPARISON OF LMS ALGORITHM DERIVATIVES USING LABVIEW FPGA

Erwin SZOPOS¹ Horia HEDESIU² Victor POPESCU¹ Lelia FESTILA¹

¹Bases of Electronics Department, ²Electrical Machines, Marketing and Management Department, Technical University of Cluj-Napoca, Romania

26-28 G. Baritiu str., Cluj-Napoca, Romania, Tel: +40264401803; Fax: +40264591340 Erwin.Szopos@bel.utcluj.ro, Horia.Hedesiu@mae.utcluj.ro, Victor.Popescu@bel.utcluj.ro, Lelia.Festila@bel.utcluj.ro

Abstract: The LMS algorithm derivatives have found wide application in many areas of adaptive signal processing and control. In this paper we study the tracking performance of the LMS derivatives in adaptive noise cancellation for a speech signal corrupted by white noise. The experiments were done for 16–tap filter lengths and as for the hardware platform the NI cRIO-9104 FPGA chassis and the NI cRIO-9014 real-time module were used. The obtained results were compared to highlight the performances of each of the LMS derivatives. The results of the experiments confirm that the NLMS variant provides the most enhanced SNR and the better tracking ability is offered by the LLMS in a 16-tap adaptive filter. The performances of the algorithms are evaluated in terms of signal-to-noise ratio (SNR) at the output of the adaptive system, tracking ability and hardware usage.

Keywords: LMS derivatives, FPGA, SNR, pipeline, tracking, hardware usage.

Number 3/2009

STUDY ON THE EFFECTS OF PRECIPITATION ON EARTH-to-SATELLITE LINKS IN THE Ka BAND

Alexandru Luigi RICOBON¹ Ramona Maria TRESTIAN²

(¹) Technical University of Cluj-Napoca, Romania; (²) Dublin City University, Ireland

Abstract: The use of higher frequency bands, above 10 GHz, for satellite communication provides a number of important benefits. It relieves congestion in the lower frequencies which are sheared with terrestrial links, in the same time providing larger bandwidths and cheaper implementation of terrestrial equipment. However, the severity of atmospheric influence on the radiowave propagation (especially due to rain) increases markedly with frequency. It is therefore necessary to identify and predict the effects of hydrometeors on wireless links.

Key words: Satellite links, Ka Band, Precipitation effects, Radar maps

A VLSI IMPLEMENTATION OF A NEW LOW VOLTAGE 4^th ORDER DIFFERENTIAL G_m-C BAND-PASS FILTERS FOR DIFFERENT APPROXIMATION IN CMOS TECHNOLOGY

Radu Gabriel BOZOMITU Vlad CEHAN

Faculty of Electronics, Telecommunications and Information Technology, “Gh. Asachi” Technical University,

Carol I No. 11 Av., 700506, Iaşi, Romania, Phone: +40-232-213737, bozomitu@etc.tuiasi.ro, vlcehan@etc.tuiasi.ro

Abstract: In this paper we present a new low voltage 4^th order differential G_m-C band-pass filter (BPF) for different approximations (Butterworth or Chebyshev), which has been designed in CMOS technology. This differential BPF is composed of two G_m-C biquad structures in cascade connection. The type of band-pass filter approximation (Butterworth or Chebyshev) can be set by an adequate choice of grounded capacitors, using digital command logic. The proposed G_m-C filter architecture is attractive for VLSI implementation and offers the possibility of tuning the centre frequency and the selectivity, using external biasing currents or modifying the grounded capacitor values. The proposed structure provides a ±15% corners variation of the centre frequency, a dynamic range of 200mV_pp(diff), distortion THD<1%, and a 10mA current consumption from a 1.8V voltage supply. The simulations performed in 65nm CMOS technology confirm the theoretical results.

Key words: Gm-C filter, biquad, transconductor, OTA, dynamic range, Butterworth, Chebyshev, CMOS, VLSI.

ON-LINE SIGNATURE RECOGNITION USING A GLOBAL FEATURES FUSION APPROACH

Eugen LUPU¹ Simina EMERICH¹ Francois Beaufort²

¹Technical University of Cluj-Napoca, Communications Dept.

26-28 Baritiu str.,Cluj-Napoca, Romania, Phone: +40-264-401266, Eugen.Lupu@com.utcluj.ro

²Ecole Supérieure d'Electronique de l'Ouest, Angers, France

Abstract: The paper presents a feature-based approach regarding on-line signature recognition, by employing the TESPAR - DZ matrices extracted from every signature. These matrices describe the different signature features shapes, such as x, y -coordinates, x, y - velocity, pressure etc. For better results, the fusion of different features was studied. Three signature databases have been employed in our experiments: DB1, DB2, our own signature databases and the Task2 database available for research purposes from the SVC2004 contest. In our work, two types of experiments were made: verification and identification (surveillance). These results are very promising taking into account the simplicity of template generation and the fact that we employed only five features in our experiments.

Key words: on-line signature, verification, identification, TESPAR, fusion, forgery, SVM.

NOVEL VOLTAGE-CONTROLLED AMPLIFIERS FOR MULTISTANDARD INTEGRATED RADIO RECEIVERS

Marius NEAG István KOVÁCS Raul ONEŢ Marina ŢOPA

Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology

26, G. Baritiu St, Cluj-Napoca, Romania, Phone: +40264 401470, Fax:+40264 591689, Marius.Neag@bel.utcluj.ro

Abstract: This paper proposes a new architecture for transconductance (Gm)-based variable-gain amplifiers (VGA) that

is better suited for multi-standard radio receivers than the classical two-Gm structure: its bandwidth and linearity are largely independent of the gain setting and both the gain range and the bandwidth can be programmed through digital controls. Unlike the classical structure - that uses two Gm cells and modifies the gain by changing the transconductance of at least one of them, usually by varying the biasing conditions - the new VGA employs three Gm cells with constant biasing; the gain variation is achieved through a differential current steering scheme. Two transistor-level implementations of the proposed VGA structure are presented, one using simple differential pairs and the other based on a more complex, highly linear Gm cell. New circuits for controlling the output common-mode voltage are proposed for both cases; these circuits do no require the placement of additional circuitry in the differential signal path. The VGAs were designed in a standard 0.18um CMOS technology to same specifications: to cover the gain range of 0dB-18dB while handling signals with amplitudes up to 800mVpkpkdiff without significant distortions; maximum bandwidth above 30MHz while driving 1pF; power consumption bellow 6mW. Simulation results show that both VGAs meet the requirements, thus validating the circuit implementations and proving the potential of the proposed architecture.

Key words: variable-gain amplifier; transconductor; differential current steering; multi-standard radio receivers.

image enhancement using a new shock filter formalism

Cosmin LUDUŞAN^{1, 2}, Olivier LAVIALLE², Sorin POP¹, Romulus TEREBE޹, Monica BORDA¹

¹Technical University of Cluj-Napoca, 26-28 Bariţiu Street, 400027, Cluj-Napoca, Romania

²Université de Bordeaux, UMR IMS 5218, 351 cours de la Libération, 33405, Talence Cedex, France

Abstract: We present a new approach based on Partial Differential Equations (PDEs) and shock filter theory for deblurring 2D Gaussian blurred images. The inherent problems of stability posed by the reverse heat equation and the discretization of shock filters are overcome by the use of only the initial conditions. This approach is not well suited for diffusion-oriented techniques since in this case the problem is ill-posed, but when dealing with multiple inputs (the case of fusion methods), it proves to provide sound results. The proposed deblurring model takes into account only the blur problem without having a noise removal component, this being a future step of our work.

Key words: partial differential equation, image fusion, shock filter, deblurring, image enhancement

A MEDICAL IMAGE FUSION METHOD FOR WEB DISTRIBUTED APPLICATIONS

Ligia CHIOREAN, Mircea F. VAIDA, Loreta SUTA

Department of Communications, Technical University of Cluj-Napoca, Cluj-Napoca, Romania

Chiorean.Ligia@com.utcluj.ro, Mircea.Vaida@com.utcluj.ro, suta_loreta@yahoo.com

Abstract: Medical images acquired with different medical modalities contain important diagnostic information. The combination of complementary data from these images can lead to more important information for diagnosis process. The fusion process allows combination of salient feature of these images. In this paper we present a method for medical image fusion, based on discrete wavelet transform. This method is implemented in a web distributed application, using Java technology. The web dedicated application allows physicians to realize a fusion from a remote place. We have also calculated some quantitative parameters for fusion process evaluation.

Key words: image fusion, wavelet transform, distributed application, multi-resolution decomposition.

Retrieving Call Detail Records from ASTERISK using SNMP

Cristian Mihai VANCEA Virgil DOBROTA

Technical University of Cluj-Napoca, Communications Department

26-28 George Baritiu Street, 400027 Cluj-Napoca, Tel: +40-264-401226, Fax: +40-264-597083

E-mails: {Mihai.Vancea,Virgil.Dobrota}@com.utcluj.ro

Abstract: Several SNMP agents could be integrated by IETF’s AgentX protocol which separates the communication with SNMP manager in two parts: AgentX between sub-agents and master agent, and SNMP between master agent and manager. The paper extends the SNMP software agent of the IP-based PBX Asterisk. The management information related to Application Layer (calls and charging) are collected in .log files, parsered and then stored into .xml files. Being an alternative to legacy commercial implementations, this solution has the advantage of using the SNMP, a generalized non-proprietary management protocol. This could be a step forward towards an integrated management where the access, distribution and core networks, together with applications are using the same mechanism which will not be external anymore as it is nowadays.

Key words: agent, Asterisk, management information base, manager, SNMP

A VLSI IMPLEMENTATION OF A NEW LOW VOLTAGE 5^th ORDER DIFFERENTIAL G_m-C BESSEL TYPE LOW-PASS FILTER WITH CONSTANT-G_m BIASING IN CMOS TECHNOLOGY

Radu Gabriel BOZOMITU, Vlad CEHAN

Faculty of Electronics, Telecommunications and Information Technology, “Gh. Asachi” Technical University,

Carol I No. 11 Av., 700506, Iaşi, Romania, Phone: +40-232-213737, bozomitu@etc.tuiasi.ro, vlcehan@etc.tuiasi.ro

Abstract: In this paper a new low voltage 5^th order G_m-C Bessel type low-pass filter (LPF) with constant-G_m biasing, designed using CMOS technology, is presented. The differential LPF is composed of two biquad structures and a first order low-pass filter. The cut-off frequency can be tuned in four steps (42 MHz, 35MHz, 27MHz and 14MHz) by modifying the grounded capacitors values, using digital command logic. In order to maintain the filter transconductance G_m independent of the process, temperature and supply voltage variations, the bias currents used by the proposed structure are made dependent on theses variations, and are given by a constant-G_m biasing source. The proposed structure provides a ±7% corners variation of the cut-off frequency, a dynamic range of 400mV_pp(diff), a distortion THD < 1%, and a 12mA current consumption from a 1.8V supply voltage. The simulations performed in 65nm CMOS technology confirm the theoretical results.

Key words: G_m-C filter, biquad, transconductor, OTA, dynamic range, Bessel, CMOS, VLSI.

BER PERFORMANCE EVALUATION OF A MOBILE WIMAX SYSTEM OVER AN ITU-R PEDESTRIAN B MULTIPATH CHANNEL

Rebeca Maria COLDA¹ Tudor PALADE¹ Emanuel PUŞCHIŢĂ¹ Irina VERMEŞAN¹ Ancuţa MOLDOVAN¹

Ştefan-Victor NICOLAESCU²

¹Technical University of Cluj-Napoca, 26-28 G.Bariţiu Street, 400027, Cluj-Napoca, Romania

Phone: +40264401403, Fax Number: +40264597083, Email:Rebeca.Colda@com.utcluj.ro

²National Communications Research Institute (INSCC), No. 6 Preciziei Street, Sector 6, Bucharest, Romania

Phone: +40213189266, Email: stnicol@co.cnscc.ro

Abstract: A realistic modeling of the radio channel’s propagation characteristics proves to be extremely important, especially when one desires to perform an accurate design of a wireless communication system. In this paper, we evaluate and optimize the performances of the ITU-R M.1225 Pedestrian B (Ped.B) channel model for a downlink (DL) mobile WiMAX system, starting from the requirements of certain multimedia services. The influence of parameters such as modulation, coding scheme and cyclic prefix (CP), was analyzed through simulations, in terms of the bit error rate (BER).

Key words: Mobile WiMAX, Ped.B channel model, modulation, coding scheme, cyclic prefix, BER.

COMBUSTION SOUND ANALYSIS TO CONTROL THE BURNING QUALITY

Eugen LUPU¹ Victor HODOR¹ Marius GHIURCAU¹

¹Technical University of Cluj-Napoca, Communications Dept.

26-28 Baritiu str.,Cluj-Napoca, Romania, Phone: +40-264-401266, Eugen.Lupu@com.utcluj.ro

Abstract: The quality of the burning process may be investigated by using different means, such as visual inspection, employing electro- chemical transducers or analyzing the sound generated by the burning process. This paper reveals attributes regarding a method to detect the different states of combustion in progress by analyzing sound analogy. These results may be then used in order to take appropriate decisions to maintain the burning process in an optimal state. The first results are very promising and encourage continuing this research in the acoustic relevance in burning quality detection. Due to the fair computational requirements of the method this may be implemented on low cost solutions like microcontrollers.

Key words: combustion quality, instability, acoustic relevance, TESPAR, archetypes.

Number 2/2009

HARDWARE STRATEGIES FOR IMAGE PROCESSING IN AN FPGA BASED MICROARRAY IMAGE PROCESSING SOC

Bogdan BELEAN Monica BORDA Albert FAZAKAS

Technical University of Cluj-Napoca

Str. Baritiu 26-28,Cluj-Napoca ; Bogdan.Belean@com.utcluj.ro

Abstract: The present paper describes a System on a Chip for microarray image processing together with the steps of a microarray experiment integrated in the proposed system. The system’s reduced size and the hardware algorithms proved to overcome the disadvantages of the existing software for microarray image processing. FPGA technology was chosen for the implementation due to its parallel computation capabilities and to the ease of reconfiguration. Hardware strategies for implementing image processing techniques are presented, together with a hardware implementation for edge detection of microarray spots using distributed memory and spatial computation.

Key words: cDNA Microarrays, FPGA, image processing, hardware algorithms.

RESOURCE MANAGEMENT IN HYBRID WIRELESS NETWORKS

USING INTELLIGENT MOBILE AGENTS

Alexandru L. RICOBON¹ Ramona M. TRESTIAN²

(¹) Technical University of Cluj-Napoca, Romania; (²) Dublin City University, Ireland

(1) alex.ricobon@com.utcluj.ro (2) ramona@eeng.dcu.ie

Abstract: Wireless networks are no more used only for data transfer and Internet access, but now also support voice and multimedia services. The tremendous demands from the users are pushing the development of mobile communications faster than ever before, leading to plenty of new advanced techniques emerging. Future wireless Internet will consist of different wireless technologies operating together. The key issue is to provide new services and maintain quality of service (QoS) in hybrid wireless networks. This can be done in many ways, one of those using mobile agents to travel across the network and configure the route or the application demands.

Key words: resource management, hybrid wireless networks, intelligent mobile agents.

OVERVIEW OF THE CROSS-LAYER PARADIGM EVOLVING TOWARDS FUTURE INTERNET

Andrei Bogdan RUS Virgil DOBROTA

Technical University of Cluj-Napoca, Communications Department, 26-28 George Baritiu Street,

Cluj-Napoca, Romania, Tel: +40-264-401226, E-mails: {bogdan.rus, virgil.dobrota}@com.utcluj.ro}

Abstract: The spectacular evolution of Internet services has requested significant improvements of the performances in communications networks. Cross-layering techniques have been selected as one of the most important issues in research, trying to change the classical OSI model and to define better mechanisms for scalability. In this paper we are investigating the main characteristics of this bi-directional vertical communication, breaking the claimed independence of physical, data link, network, transport, and application layers. Some deployment scenarios described herein could help the future Internet designers to follow a clean slate approach.

Key words: cross-layer, Future Internet, QoS

DIGITAL IMPLEMENTATION OF THE SIGMOID FUNCTION FOR FPGA CIRCUITS

Alin TISAN, Stefan ONIGA, Daniel MIC, Attila BUCHMAN

Electronic and Computer Department, North University of Baia Mare, Baia Mare, Romania

Str. Dr. Victor Babeş, nr. 62A, tel. 0362-401265 ext. 234, alin.tisan@ubm.ro

Abstract: In this paper, is proposed a method to implement in FPGA (Field Programmable Gate Array) circuits different approximation of the sigmoid function. Three previously published piecewise linear and one piecewise second-order approximation are analyzed from point of view of hardware resources utilization, induced errors caused by the approximation function and bits representation, power consumption and speed processing. The major benefit of the proposed method resides in the possibility to design neural networks by means of predefined block systems created in System Generator environment and the possibility to create a higher level design tools used to implement neural networks in logical circuits.

Key words: Sigmoid approximation, FPGA implementation, System Generator

SECRET WRITING BY DNA HYBRIDIZATION

Monica E. BORDA Olga TORNEA Tatiana HODOROGEA

Technical University of Cluj-Napoca, Communications Department

Gh. Baritiu str. 25-27, fax: 004-64-401575, Romania, E-mail: Monica.Borda@com.utcluj.ro

Abstract: This paper presents an original algorithm of secret writing by DNA hybridization, based on existing ideas described in the literature. In order to perform highly secure communication, the cryptographic, steganographic and biomolecular computation strength is used. The proposed algorithm is exemplified using matlab bioinformatics toolbox.

Key words: Secret writing, DNA hybridization, PCR, DNA cryptography, steganography, biomolecular computation

Acoustic ModelLINg and OPTIMization of A Room

Norbert TOMA, Marina TOPA, Irina MURESAN, Botond KIREI, Marius NEAG, Albert FAZAKAS

Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca

Str. G. Bariţiu nr. 26-28,400020 Cluj-Napoca, Romania, Tel: +40264401243; Fax: +40264591340 Marina.Topa@bel.utcluj.ro

Abstract: The paper presents the analysis and improvement of a room’s acoustics using both software and hardware tools. First a model of the room is proposed taking into account the comparison of simulation and experimental results. Next improvements in the architecture are suggested and then the optimal placement of loudspeakers and listeners is provided. The performance is measured in terms of acoustic parameters.

Key words: room acoustics, modelling, reverberation time, sound pressure, acoustic parameters.

SIMULINK LIBRARY OF BASIC BUILDING BLOCKS FOR TERNARY LOGIC

Emilia SIPOS, Gabriel OLTEAN, Costin MIRON

Bases of Electronics Department, Technical University of Cluj-Napoca

26-28 G. Baritiu str., Cluj-Napoca, Romania, Tel: +40264401470; Fax: +40264591340;

Emilia.Sipos@bel.utcluj.ro

Abstract: A library of basic building blocks implemented in Simulink for ternary logic is presented. The basic blocks are: inverters, minimum, negated-minimum, maximum and negated-maximum. The considered ternary logic is unbalanced ones, with 0, 1 and 2 logic levels. With basic building blocks we implement the ternary D flip-flap-flop, with binary and ternary clock. The simulation results validate the correct operation for all implemented building blocks.

Keywords: ternary inverters, minim and maxim ternary circuits, D flip-flap-flop, Simulink block

RELEVANT CONTEXT DISCOVERY FOR PERVASIVE SERVICES INVOLVING USER CONTROL

Marcel CREMENE¹, Alin DRIMUS³, Jean-Yves TIGLI², Stephane LAVIROTTE², Gaëtan REY², Michel RIVEILL², Costin MIRON¹, Mircea VAIDA¹

¹Technical University of Cluj-Napoca, Communications, {cremene, vaida}@com.utcluj.ro,miron@bel.utcluj.ro

²University of Nice Sophia-Antipolis, Ecole Polytechnique, {tigli, lavirott, gaetan.rey, riveill}@unice.fr

³University of Southern, SDU, Denmark,drimus@mci.sdu.dk

Abstract: Dynamic context adaptation is a central concept in Pervasive and Ubiquitous Computing. In this paper we have analyzed the problem of context relevance and we have chosen an approach where the user is also involved also in the relevance control. The advantage of involving the user is motivated by the fact that a completely automatic context management cannot deal with unpredicted situations. The proposed solution was implemented on the top of WComp middleware. An easy to use interface let the user the possibility to easily manage its context.

Key words: Ubiquitous computing, Relevant context, Context discovery, Intelligent Devices.

MAXIMUM POWER POINT TRACKING SIMULATOR IN CHARGING PHOTOVOLTAIC SYSTEMS

Dorin M. PETREUŞ Ionuţ C. CIOCAN Adina R. RUSU Dorin V. CADAR

Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology,

26-28 G. Bariţiu Street, Cluj-Napoca, Romania, phone: 0264 401 412, email: Dorin.Petreus@ael.utcluj.ro

Abstract: The task of Maximum Power Point Tracking (MPPT) is to operate with photovoltaic (PV) modules in a manner that allows them to produce all the power they are capable of. In this paper a system for modeling different MPPT algorithms is presented. The functionality of the model proposed was validated through the comparison of the simulation results with the ones experimentally obtained from the implemented evaluation board. Different algorithms were developed in C language using Keil software 8051 development tool, µVision. The maximization of the output power of PV system is controlled through a 8051 microcontroller. The entire PV system was simulated using Proteus 7 Professional and can be used to test any MPPT algorithm.

Key words: MPPT algorithms, PV module, macro-model.

Number 1/2009

Fiber segmentation in composite materials using marked point processes

Barna KERESZTES^1,2, Olivier LAVIALLE¹, Sorin POP², Monica BORDA²

(1) Université de Bordeaux, LASIS - IMS – 351, Crs. de la Libération, 33405 Talence Cedex

(2) Technical University of Cluj-Napoca 15 Daicoviciu Street, 400020 Cluj-Napoca, Romania

e-mail: barna.keresztes@ims-bordeaux.fr tel: +33 540003624

Abstract: This paper presents a new method dedicated to unsupervised 2D segmentation of fibers in a section of composite carbon-fiber materials. The framework relies on a marked point process algorithm. We shall create random elliptical objects to fit the fiber distribution in the image. The interaction rules between the objects complete the model. Using a Markov Chain Monte Carlo (MCMC) method, the algorithm converges to a configuration which is close to the fiber distribution in the images. At each step, the configuration is evaluated considering its proximity to the target distribution. In order to achieve this task, we propose a mixed data model using both grey level values and gradients to evaluate the likelihood of the current configuration. This mixed model overcomes the problems of luminance variation, contour discontinuities and high noise level. Finally, the results on the composite material sections illustrate the efficiency of the segmentation and suggest that the marked point processes can be a promising tool for fiber detection.

Keywords: marked point process, composite materials, MCMC methods

DIVIDE-AND-CONQUER PIECEWISE LINEAR APPROXIMATION OF GAIN AND PHASE EVALUATION

Lacrimioara GRAMA¹ Corneliu RUSU²

1;2Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology,26-28 G. Baritiu Street, Tel. +40 264 401804, Fax. +40 264 591340,

1 E-mail: Lacrimioara.Grama@bel.utcluj.ro 2 E-mail: Corneliu.Rusu@bel.utcluj.ro

Abstract: Certain approaches to phase approximation ask as a first step in implementation for a piecewise linear fitting of gain. When the breakpoints are given, there are few ways to determine the slopes of the broken line approximation. In this paper we determine the breakpoints using a divide-and-conquer approach. Results to piecewise linear approximation of gain and of the resulting phase approximation are presented. For this purpose, we consider certain frequency responses, then the gain data are corrupted or altered.

Keywords: piecewise linear approximation, divide-and-conquer, phase approximation, Hilbert transform

3D SURFACE RECONSTRUCTION AND COMPARISON SYSTEM USED FOR MARKING LOCAL DEFECTS OF TESTED OBJECTS

Adrian CHIOREANU, Aurel VLAICU, Bogdan ORZA

Technical University of Cluj-Napoca

adrian.chioreanu@com.utcluj.ro, aurel.vlaicu@com.utcluj.ro, bogdan.oraza@com.utcluj.ro

Abstract: Nowadays automatic 3D acquisition systems allow building accurate models of real 3D objects in a cost- and time-effective manner. This technology was used in a particular application context: shape defects detection. Specific needs of this domain are: medium-high accuracy, reliability, and a short acquisition time. A low-cost 3D reconstruction and testing prototype system was designed, based on structured light which compares the reconstructed shape of the tested objects with the reconstructed shape of a reference object. The scanner architecture and the first results obtained by implementing a prototype used to mark local defects of tested objects are presented.

Key words: 3D reconstruction, quality testing, structured light, defects marking.

STUDY ON THE PERFORMANCE OF CODING-ASSISTED SWARM COMMUNICATIONS

Anghel BOTOŞ , Mihai P. ŞTEF, Aurel VLAICU

Technical University of Cluj-Napoca

15 C. Daicoviciu str., 400020 Cluj-Napoca, Romania, Tel: +40264401906, Anghel.Botos@com.utcluj.ro

Abstract: Swarm communications represent some of the main solutions proposed for large content distribution in the current and future data communication networks. One of the possible solutions to improve the performance of swarm communication is represented by the use of Digital Fountain and Network Coding techniques integrated in the data exchange between swarm peers. The paper proposes a comparative study of the non-coded and coding assisted swarm communications in different settings related to the amount of employed signaling, supersymbol size, supersymbol number, and number of initial seeds. The study is based on computer simulations performed in some particular network topology and the goal is to identify the settings of coding assisted swarm communications which ensure significant decrease of its content distribution time.

Key words: Content distribution, Digital Fountain, Network Coding, swarm communications.

PERFORMANCE ANALYSIS OF NETWORK CODING-BASED COOPERATION ALGORITHMS EMPLOYED IN CELLULAR NETWORKS

Mihai P. ŞTEF, Anghel BOTOŞ, Aurel VLAICU

Technical University of Cluj Napoca

15, Constantin Daicoviciu, Cluj Napoca, Romania,Tel. +40264401906 , Mihai.Stef@com.utcluj.ro

Abstract: The employment of the network coding, combined with distributed FEC coding, in cooperative transmission schemes leads to a more efficient use of the additional time-frequency resources required by cooperation, i.e. by allowing a single relay node to serve two or more user terminals. This paper proposes a low complexity cooperation algorithm which employs network and distributed channel coding techniques in a two sources – one relay node cooperation scheme for the uplink connection of a cooperative cellular network. It also, provides a description and an analysis of the “mother” algorithm, Separate Network and Channel Coding (SNCC). The performance of the proposed cooperation algorithm is evaluated in terms of BER and PER in several significant scenarios. The performance provided by this algorithm is compared to the performance of non-cooperative coded transmission and to the performance of the “mother” algorithm, the SNCC algorithm.

Key words: cooperative diversity, channel coding, network coding, turbo codes

TTS EXPERIMENTS: ROMANIAN PROSODY

Arpad Zsolt BODO^{*, **}, Gavril TODEREAN^*, Ovidiu BUZA^*

^*Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology;
^**Sprint Consulting Ltd.,Cluj-Napoca, Romania

zsolt.bodo@sprintconsulting.com

Abstract: In a text-to-speech synthesis system, prosody prediction and generation, is one of the most critical tasks. Prosody is composed by the melody, intensity and by the rhythm of speech. The current paper focuses on the Romanian intonation prediction and generation within our TTS system. Experimental results achieved by the development of different prosodic modules are presented in this work. The process is followed from text pre-processing, sentence type determination, through automatic labeling of sentences, till the labeling and transposing of the intonation curves. The creation and storage of relative intonation curves, as the developed different rule systems described in XML are also topic of this work.

Key words: Intonation; Melody; Labeling; Prosody; Sentence type; Text analysis; Intonation curve; XML rules.

VECTOR CONTROL OF INDUCTION MOTOR BASED SPACE VECTOR MODULATION

L. BOURAS M. KADJOUDJ

Batna University, Algeria, BP 336 Batna, email: kadjoudj_m@yahoo.fr

Abstract: It is well known that the vector control based on hysteresis regulators of induction machine presents less performance in term of current quality witch translated by strong ripples at steady state what increases the commutation losses in the inverter. For these reasons, we propose a new approach of FOC based on the SVM technique. Simulation results are shown to demonstrate the validity of the proposed methods.

Key words: Induction Motor, FOC, SVM, Hysteresis.

FUZZY LOGIC BASED DECISION MAKING FOR HYDROELECTRIC ENERGY GENERATION IN A CASCADED HYDROPOWERPLANTS

Gabriel OLTEAN*, Mihaela GORDAN*, Emilia ŞIPOŞ*, Eugen STĂNCEL**, Ovidiu DANCEA**

*Bases of Electronics Department, Technical University of Cluj-Napoca

26-28 G. Baritiu str., Cluj-Napoca, Romania, Tel: +40264401470; Fax: +40264591340;

**IPA SA Cifaat Cluj, Zorilor 15, Cluj-Napoca, Romania Tel: +40264-596155,

Gabriel.Oltean@bel.utcluj.ro

Abstract: Hydroelectric energy is a major source of renewable electricity in the world. The industry is continuously searching for ways to improve the efficiency and reliability with which it produces energy. A method based on subtractive clustering of historical data, to develop fuzzy rule-based systems for optimal hydroelectric energy generation allocation for a hydropowerplant is proposed in this paper. The applicability of this modeling approach is demonstrated by a study on the cascaded hydropowerplants on Somes River in Romania. The optimal planning of energy generation is determined on total demand of energy production for entire hydropowerplants cascade, water storage stage, and reservoir inflows. Two fuzzy systems, FEA1, encapsulating information only about the hydropowerplant under consideration, and FEA2, encapsulating information about all the cascaded hydropowerplants, are developed. Both fuzzy models was able to model, with good results, the optimal operation of the considered hydropowerplant, with a plus for FEA2 that provides more accurate results in the testing data set.

Keywords: hydroelectric energy generation, hydro power plant, fuzzy rule-based system, subtractive clustering.

Number 4/2008

DIGITAL IMPLEMENTATION OF ARTIFICIAL REVERBERATION ALGORITHMS

Irina DORNEAN, Marina ŢOPA, Botond Sandor KIREI - Faculty of Electronics, Telecommunications and Information Technology, Technical University of Cluj-Napoca , Str. G. Bariţiu nr. 26-28,400020 Cluj-Napoca, Romania, Tel: +40264401470; Fax: +40264591340 Irina.Dornean@bel.utcluj.ro

Abstract : The paper presents the digital implementation of signal processing algorithms that simulate natural concert hall reverberation. It deals with the complete artificial reverberators proposed by Schroeder, Moorer and Gardner. For each reverberation algorithm the implementation was done in two steps: 1. the algorithm is checked using a model in Matlab Simulink; 2. the Verilog code is written and tested. The results prove that the Verilog design is feasible and can be further developed for acoustic improvements of rooms.

Key Words: : artificial reverberators, FPGA

A FUNCTIONAL MODEL FOR A STANDALONE FINGERPRINT RECOGNITION SYSTEM

Liviu ANDRON, Toader MELINTE , Seektron SRL, andron@seektron.ro

Abstract: The paper describes a functional model for a standalone fingerprint recognition system, representing a minimum implementation to demonstrate the functionality of such system. For simplicity, the model uses only a fingerprint recognition module and a demonstration board with a PIC microcontroller. The development tools were that common used by the PIC microcontroller family.

Key Words: fingerprint recognition system, fingerprint recognition module, embedded system, microcontroller.

FEEDBACK CONTROL VS. INTEGRATED CONTROL.A CASE STUDY: VARIABLE VOLUME ROOM EXPERIMENTAL BENCHMARK

Adrian JOLDIŞ, Dorin ISOC, Ion GAZIUC, Aurelian IGNAT-COMAN - Technical University, Romania 400020, Cluj-Napoca str.C.Daicoviciu nr.15 , adrian.joldis@com.utcluj.ro, dorin.isoc@aut.utcluj.ro, ion.gaziuc@yahoo.com, aurelian.ignat@aut.utcluj

Abstract: Systems are in continuous diversification of their application domains and functioning specification requirements. Now, control systems’ development is mainly oriented towards completing of conventional feedback control with new techniques and design means. This work deals with integrated control problems compared to feedback control systems. The purpose is to extract defining characteristics of feedback and integrated control that emphasize the main differences between these two terms and manners to understand the control. Variable volume room experimental benchmark is introduced together by means a functional specification. Some design possibilities are introduced to demonstrate the general discussed aspects based on a real example.

Key words: integrated control, systems design, feedback control, experimental benchmark.

AN EXPERIMENTAL ARCHITECTURE FOR BASIC IPTV CONCEPTS IMPLEMENTATION AND TESTING

Radu ARSINTE - Technical University Cluj-Napoca, Tel: +40-264-595699, Str. Baritiu 26-28, Radu.Arsinte@com.utcluj.ro

Abstract: The paper describes basic concepts and a practical implementation of a simple architecture for IPTV content creation and distribution. The architecture is conceived to make possible extensive tests. Audio-video content is acquired using different conventional technologies: DVB-S, CATV Headend, video streaming servers. This experimental network is intended to be used for throughput, QoS and audio-video quality measurements. Some of the most relevant features and tests (still in progress) are described in the final part of the paper.

Key words: IPTV, Network, Audio-video streaming, Content management

BROADBAND POWER AMPLIFIER NONLINEARITY CANCELLATION IN OFDM SYSTEMS

Stefan WERNER, Fernando GREGORIO, Juan COUSSEAU, Jose FIGUEROA,, Risto WICHMAN - Helsinki University of Technology P.O. Box 3000, 02015 TKK, Finland, , Universidad Nacional del SurAv. Alem 1253, Bah´ıa Blanca, 8000 Argentina

Abstract: One drawback of OFDM systems is the high peak-to-average power ratio, which imposes strong requirements on the linearity of power amplifiers (PAs). Such linearity requirements translate into high back-off that results in low power efficiency. In order to improve power efficiency, a PA nonlinearity cancellation (PANC) technique is introduced which reduces the nonlinear distortion effects on the received signal. In order to deal with broadband high-power PA (HPA) distortion, the PANC technique employs aWiener-Hammerstein model. Simulation results show that good levels of distortion cancellation are possible for a system with a nonlinear HPA with memory.

Keywords: Orthogonal frequency-division multiplexing (OFDM), nonlinear distortion, power amplifier (PA) nonlinearities cancellation, Wiener-Hammerstein model.

MULTIPLE-INPUT ADAPTIVE COMBINER-EQUALIZER FOR WIRELESS COMMUNICATIONS

Ligia CREMENE Marcel CREMENE Nicolae CRISAN Tudor PALADE, Technical University of Cluj-Napoca, 15 Daicoviciu street, 0264-401914, Ligia.Cremene@com.utcluj.ro

Abstract: Receiver diversity is one of the most efficient techniques against fading and interference effects. MRC (Maximum Ratio Combining) was proven to ensure the highest diversity gain, but it is also the most complex compared to other combining techniques. In this paper we propose a multiple-input adaptive combiner-equalizer that provides better performance at lower complexity. The novelty of the solution lies in the unified combining-equalization approach - the two classical operations being performed simultaneously and not sequentially. A first set of simulations has shown that the performance of the proposed combiner-equalizer is superior to that of a classical MRC implementation.

Key words: Adaptive Equalization, Receiver Diversity, MRC (Maximum Ratio Combining).

THE IMPROVEMENT OF THE WAVELET ANALYSIS TECHNIQUES BY USING B-SPLINE FUNCTIONS FAMILY

Simina EMERICH - Communications Department, Technical University of Cluj-Napoca, G. Baritiu 26-28, 400027, Cluj-Napoca, Romania, Simina.Emerich@com.utcluj.ro

Abstract: The goal of this paper is to justify the use of fractional B-spline bases in several Wavelet Analysis Techniques. We are interested in better controlling the differentiation behavior of the wavelet. For that purpose, we propose an edge detector algorithm by using the wavelet transform local maxima, based on B-spline functions. Another task is to search the connection available between energetic analysis and the chosen type of wavelet family and to use the results for Glottal Closure Instants Estimation in Voiced Speech.

Key words: Wavelets, fractional B-splines, differentiator, energetic analysis

BROADBAND WIRELESS ACCESS AND MOBILITY SOLUTION USING IEEE 802.16: CONCEPTS AND ARCHITECTURE

Rebeca Maria COLDA , Tudor PALADE, Emanuel PUŞCHIŢĂ - Faculty of Electronics, Telecommunications and Information Technology, Communications Dept. , 26 G.Baritiu Street, 400027 Cluj-Napoca, Romania, E-mail rebeca_colda@yahoo.co.uk

Abstract: WiMAX (Worldwide Interoperability for Microwave Access) is a wireless broadband access solution, developed according with the IEEE 802.16 standard specifications with the purpose of delivering high speed radio access, most commonly Internet access, especially for those areas where a conventional solution, based on cable or DSL, proves to be ineffective. The early WiMAX solutions were based on IEEE 802.16-2004 and targeted fixed applications. In December 2005, the IEEE group completed and approved IFEE 802.16e-2005, an amendment to the IEEE 802.16-2004 standard that added mobility support.

Key words: WiMAX, IEEE 802.16-2004, IEEE 802.16e-2005, BS, SS.

PERFORMANCE ANALYSIS OF MIMO TECHNOLOGY IN MOBILE WIRELESS SYSTEMS

Ancuţa MOLDOVAN, Traian VILA, Tudor PALADE, Emanuel PUŞCHIŢǍ – Technical University of Cluj-Napoca,26-28 G. Bariţiu Street, 40027, Cluj-Napoca, Romania, phone:40264401914,e-mail:Ancuta.Moldovan@com.utcluj.ro

Abstract: The main goals in developing new wireless communication systems are increasing the transmission capacity and improving the spectrum efficiency. Space-time codes and spatial multiplexing codes exploit the richly scattered wireless channel by using multiple transmit and receive antennas in order to improve the reliability of data transfer and greatly increase the data transmission rates without additional radio resource requirements. Significant performance gain is achieved by varying some transmit and receive parameters.

Key words: MIMO, space-time coding, space division multiplexing

DESIGN OF AN INTELLIGENT SYSTEM FOR MONITORING AND ASSISTING OF NEOPLAZIC PATIENTS

Corneliu RUSU, Sergiu CADAR, Silviu CORBU, Lacrimioara GRAMA, Marius SIRBU - 1Universitatea Tehnica Cluj-Napoca, Facultatea de Electronica, Telecomunicatii si Tehnologia Informatiei Baritiu 26-28, Tel. +40 264 401804, Fax. +40 264 591340, E-mail: Corneliu.Rusu@bel.utcluj.ro, 2Institutul de Cercetari pentru Instrumentatie Analitica Cluj-Napoca Donath 67, Tel. +40 264 420590, Fax. +40 264 420667, E-mail: sergiu.cadar@icia.ro 3Universitatea din Oradea, Facultatea de Medicina Universitatii 1, Tel. +40 259 408105, Fax. +40 259 432789, E-mail: smcorbu.oncology@yahoo.com 4Universitatea Tehnica Cluj-Napoca, Facultatea de Electronica, Telecomunicatii si Tehnologia Informatiei Baritiu 26-28, Tel. +40 264 401804, Fax. +40 264 591340, E-mail: Lacrimioara.Grama@bel.utcluj.ro 5SC WS COMM SRL Cluj-Napoca, Republicii 107, Tel./Fax. +40 364 440120, E-mail: marius@wscomm.ro

Abstract: The aim of this work is to create an intelligent client-server structure through which the neoplazic persons that use the opioid medicine to be monitorized and assisted at home, and also to efficiently intervene to lend the essential assistance. The patient has at his/her disposal an interface which allows the transmission of some information regarding his/her healthiness through a reception server. This information is analyzed in accordance with some treatment criterions and protocols well regulated, and decisions for adjusting the doses of the opioid medicine are taken function of the intensity of the algic symptomatology.

Keywords: Intelligent client-server system, oncology, biometry, opioid medication.

Number 3/2008

ADVANCED FLEXIBLE ADAPTIVE FILTERING – TOWARDS STRUCTURE, TIME AND STATISTICAL AGILITY

Colin F.N. COWAN - ECIT, Queen’s University Belfast, Northern Ireland Science Park, Queen’s Island, Belfast BT3 9DT. Phone:+44 (0)28 9097 1879, Fax:+44 (0)289097 1702, Email: c.cowan@ecit.qub.ac.uk

Abstract: This paper considers the evolution of modern adaptive filtering in the context of applications in wireless communications. The main contribution is to introduce a common framework which potentially allows the adaptive system to include structure, e.g. number of coefficients, as one of the adaptable parameters. The proposed structure also allows for statistical variations to be taken into account in a highly time variant manor. Some examples are also cited to illustrate these features.

Key Words: Adaptive filters, structure adaptation, Kurtosis, time variation, equalization, echo cancellation.

SAFE PRE-PASS SOFTWARE BYPASSING FOR TRANSPORT TRIGGERED PROCESSORS

Pertti KELLOMAKI, Vladimir GUZMA, Jarmo TAKALA - Tampere University of Technology, Finland , P.O. Box 553, FIN-33101 Tampere, Finland Fax +358 3 3115 4561. Email {firstname.lastname}@tut.fi

Abstract: The Transport Triggered Architecture is an architectural style where the internal transport buses of a processor are exposed in the instruction set. Since software has full control over data transports, data can be directly transfered between functional units, an optimization known as software bypassing. Software bypassing can potentially reduce the need for general purpose registers, because temporary values can be kept in functional units. Taking advantage of the reduced register pressure requires software bypassing to be done before register allocation and scheduling. In this paper we analyze under which conditions software bypassing can be done on a Data Dependence Graph without compromising the schedulability of the graph.

Key Words: Triggered Architecture, Software, Bypassing

MULTY BINARY TURBO CODED WOFDM PERFORMANCE IN FLAT RAYLEIGH FADING CHANNELS

Marius OLTEAN, Maria KOVACI, Horia BALTA, Andrei CAMPEANU - Faculty of Electronics and Telecommunications, Timisoara, Romania Bd. V. Parvan, nr.2, Timisoara, Phone: 0256403307, email: marius.oltean@etc.upt.ro

Abstract: In this paper, the application of Multi Binary Turbo Codes (MBTCs) to two multicarrier modulation techniques used for transmission in the flat radio fading channel is studied. Thus, the performance of the coded versions of Orthogonal Frequency Division Multiplexing (OFDM) and of its wavelets based version (WOFDM) is evaluated under several simulation scenarios. These scenarios take into account the time variability of the radio channel, by the mean of the Doppler shift parameter. An important conclusion is that, under the assumptions made in this paper, WOFDM outperforms OFDM. By the other hand, this technique shows little influence on the choice of the wavelets mother used in the modulator.

Key words: Duo binary Turbo Codes, WOFDM, OFDM,  Flat Fading

A STUDY OF THE PERMUTATION SCHEMES USED IN THE MOBILE WIMAX

Horia BALTA¹, Daniel BOSNEAGU², Maria KOVACI¹, Marius OLTEAN ^{1,2 1} Electronics and Communications Faculty, Politehnica University, Bd. V. Parvan, Timisoara, 40256403307, (horia.balta, maria.kovaci, marius.oltean)@etc.upt.ro ²ALCATEL - LUCENT Timisoara

Abstract: With the goal of decreasing the interference at sub-carrier level, the allocation of sub-carrier frequencies in an OFDMA based system is selective. This selective allocation can be seen like a permutation of a frequency “sequence”. This paper proposes a new analyzing tool for the frequencies permutation used in WiMAX mobile communications: the distances spectrum (DS). This tool can offer a meaningful measure for these permutations, with respect to the most performing permutations used in turbo-codes, which, given their major importance, are strongly studied and developed.

Key words: permutations, sub-carrier frequency, interleaving distances spectrum, spreading degree.

WATERMARKING BASED ON THE HYPERANALYTIC WAVELET TRANSFORM

Corina NAFORNITA, Alexandru ISAR - Electronics and Communications Faculty, Politehnica University,

2 Bd. V. Parvan, Timisoara, 40256403307, (corina.nafornita, alexandru.isar)@etc.upt.ro

Abstract: In this paper we present a new technique for watermarking in the hyperanalytic wavelet domain. This has the advantage of better directional selectivity and shift invariance compared to the classical wavelet filter. The watermark is inserted into all levels of decomposition using a perceptual mask conceived by the authors. We compare this technique with two systems in the DWT domain, and show its superiority. We study the resistance of the three watermarking systems against various signal processing attacks. All three methods successfully detect the watermark. This leads to the conclusion that the Hyperanalytic Wavelet Transform (HWT) increases the payload and robustness.

Key words: HWT, watermarking, perceptual mask, capacity.

ON THE TUNING PERFORMANCE OF AN ACTIVE RF BANDPASS FILTER

Cristian ANDRIESEI, Liviu GORAŞ - “Gh. Asachi” Technical University, Iaşi 700506 Iasi, Bd. Carol I, No. 11, Phone: (+40) 232 270041, Fax: (+40) 232 217720, lgoras@etc.tuiasi.ro

Abstract: The paper describes a new principle of independent frequency and quality factor tuning suitable for simulated  inductors based active RF filters. The method makes use of two dc decoupled negative resistances in order to avoid frequency and quality factor inter–dependency. A differential architecture is studied and several considerations regarding the tuning capability and stability behavior are presented. The target application is a multi–standard active bandpass filter envisaging wireless transceivers.

Key words: simulated inductor, CMOS RF filters, negative resistance, independent tuning.

A ZIGBEE SOLUTION FOR TELEMEDICINE APPLICATIONS

Adriana N. SÎRBU, Victor-Andrei V.D. MAIORESCU, Ioan I.CLEJU - Faculty of Electronics & Telecommunications, Technical University “Gh. Asachi” Iaşi B-dul Carol I, no. 11, 700506 Iaşi, Romania, Phone +4-232213737/259,Fax+4-232217720,asirbu@etc.tuiasi.ro

Abstract: The paper addresses several aspects regarding wireless communications in telemedicine applications. Possible solutions and comparisons between different technologies are presented.  A ZigBee based implementation case study is commented.

Key words: telemedicine, wireless technologies, ZigBee, sensor network, wireless body area network.

MULTIMODAL BIOMETRIC SYSTEMS OVERVIEW

Eugen LUPU, Petre G. POP - Technical University of Cluj-Napoca phone: +40-264-401-266; fax: +40-264-556-244 Email: Eugen.Lupu@com.utcluj.ro

Abstract: Biometric systems allow automatic person recognition based on physical or behavioral features which belong to a certain person. Each biometric feature has its limits and no biometric system is perfect so unimodal biometric systems raise a variety of problems. To over fulfilling some of the mentioned inconvenient and limitations and to increase the level of security the multimodal biometric systems are used. This paper discusses the main features of the multimodal biometric system: architecture, level of fusion, methodology used for integrating the multiple verifiers and normalization techniques. The main applications of the multimodal biometric systems are also presented.

Key words: biometric, authentication, multimodal, security, access control, fusion. normalization

EXPLORING FINE PHONETIC DETAIL FOR ROMANIAN TEXT TO SPEECH SYNTHESIS

Mircea GIURGIU - Technical University of Cluj-Napoca, Telecommunications Department, 26 Baritiu Str., 400027 Cluj-Napoca, Romania, E-mail: Mircea.Giurgiu@com.utcluj.ro

Abstract: This paper presents several experimental results on the study of Fine Phonetic Detail (FPD) in order to create the basis of a computational model for prosody description in a system for Text To Speech (TTS) synthesis. Different manifestation levels of FPDs together with their importance for TTS are presented. The experiments reported here, mainly take into consideration the linguistic aspect of prosody in a qualitative and quantitative manner: the influence of the accent at word and sentence level, the intonation, the rhythm and the speech rate.

Key words: speech processing, fine phonetic details, prosody control, text to speech synthesis

THE ROLE OF PROGRAMMABLE DIGITAL SIGNAL PROCESSORS (DSP) FOR 3G MOBILE COMMUNICATION SYSTEMS

Sorin ZOICAN - Politehnica University of Bucharest , Bdul Iuliu Maniu nr. 1-3, Phone: +4021 402 48 03, Fax: +4021 316 96 32, Email: sorin@elcom.pub.ro

Abstract: This paper illustrated the role of Digital Signal Processors (DSP) for third generation mobile systems. The currently deployed wireless infrastructure supports the third generation of mobile communications comprised of many distinctly different standards. As a result, interoperability among these standards is nearly impossible with respect to the cellular base station.  Many solutions for base station or mobile station have been implemented over the years, and each solution required a combination of two components, ASICs (Application Specific Integrated Circuits), and DSPs (Digital Signal Processors). This two-chip solution partitions the processing tasks between the ASIC and DSP, respectively. Although this solution is functionally acceptable, its system cost and flexibility are not completely optimized. Global objectives and attributes that include worldwide roaming, universal connectivity, high data transmission rates, location service capability, and support for high-quality multimedia services are now required. The paper is organized as follows: the 3G requirements, advanced technologies that involve DSP, architecture examples and conclusions.  

Key words: Digital Signal Processors, 3G Mobile Communications Systems

CAMERA PHONE BASED BARCODE DECODING SYSTEM

Romulus TEREBES⁽¹⁾,Otmane LAHLOU^{(2) - (1)}The Technical University of Cluj-Napoca, Romania, 15 C. Daicoviciu Street,Cluj-Napoca, Romania, ⁽²⁾ENSEIRB, France 1 av. du docteur Schweitzer,33402, Talence cedex, France

Abstract: We propose a 1D barcode acquisition and decoding system based on the use of J2ME enabled mobile phones. The approach relies on image processing techniques to correct the distortions introduces by the acquisition device, segments the useful information, decodes it and subsequently sends it to a web server for further processing.

Key words: Image processing, 1D barcodes, Java 2 MicroEdition.

Number 2/2008

AUDIO SUSCEPTIBILITY OF THE BUCK CONVERTER IN CURRENT-MODE POWER STAGE

Costel PETREA - Technical University “Gh.Asachi” Iasi Carol I, no.11 , 700506, cpetrea@etc.tuiasi.ro

Abstract: For the Buck Converter operating in Current-Mode Power Stage, the subharmonic oscillations can be eliminated by adding an external ramp. This ramp leads to reaching the null value of the audio susceptibility. When the null value of the audio susceptibility is reached, the error signal phase is shifted with –180 degrees . The external ramp value chosen for the null value of the audio susceptibility does not always coincide with the value for eliminating the subharmonic oscillations. Our paper deals with the compromise in order to obtain a power supply with an audio susceptibility as close to the null value as possible, a fast transient response as well as a good stability.

Key words: Buck Converter, Audio susceptibility, subharmonic oscillations, external ramp.

A VISUAL TOOL FOR DNA REPEATS LOCALIZATION

Petre G. POP - Technical University, Cluj-Napoca, Comm. Dept.,E-mail:Petre.Pop@com.utcluj.ro

Abstract: The detection of tandem repeats is important in biology and medicine as it can be used for phylogenic studies and disease diagnosis. A major difficulty in identification of repeats arises from the fact that the repeat units can be either exact or imperfect, in tandem or dispersed, and of unspecified length. This paper presents results obtained by combining grey level spectrograms with a novel numerical representation to isolate position and length of tandem repeats (TRs) in DNA sequences.

Key words: Genomic Signal Processing, Fourier analysis, Spectrograms, Sequence Repeats.

PHASE UNWRAPPING ALGORITHMS IMPLEMENTED ON A SYMBIAN BASED MOBILE DEVICE

Carol RUS, Lacrimioara GRAMA, Andrei DUSLEAG, Corneliu RUSU - Technical University of Cluj-Napoca, Romania , Email: Lacrimioara.Grama@bel.utcluj.ro

Abstract: Phase unwrapping is a procedure that allows 3D representation for surfaces when the data is acquired using interferometry based techniques. An interferogram contains phase difference values wrapped onto a fixed range of angles 0-360 degrees. In order to compute surface heights and generate an elevation model, the interferogram have to be unwrapped. This paper presents two 2D phase unwrapping algorithms: the zig-zag approach and the least-squares method, and shows an implementation of them on a mobile device based on Symbian S60 platform. Also, filtering methods (i.e mean filter and vector filter) for preprocessing the interferogram are discussed and implemented. The implementation was tested using SAR (Synthetic Aperture Radar) interferograms. Based on the results, it can be concluded that mobile devices offer enough processing and graphical power and development options, in order to be a good choice to test some new signal processing algorithms.

Key words: Interferogram, DEM, Phase Unwrapping, Symbian, S60.

FOUR QUADRANT ANALOG CURRENT MODE MODULAR MULTIPLIER DESIGNED WITH DIFFERENTIAL AMPLIFIERS

Lorant Andras SZOLGA, Lelia FESTILA, Mihaela CIRLUGEA - Basis of Electronics Department, Faculty for Electronics and Telecommunications, Technical University of Cluj-Napoca,26. Baritiu Street, 400027, Romania,Tel/fax:0040-264-591340, lorant.szolga@bel.utcluj.ro

Abstract: In this paper we propose two and four quadrant analog current mode multipliers, implemented with bipolar differential amplifiers. The simplicity, low power consumption and high frequency operation recommend them for real time applications.

Key words: analog multilplier, hyperbolic tangent domain, FF-1functions

WAVELETS AS FEATURES FOR OBJECTS RECOGNITION

Anca APATEAN (DISCANT), Alexandrina ROGOZAN, Simina EMERICH, Abdelaziz BENSRHAIR- Tehnical University of Cluj-Napoca, Cluj-Napoca, Romania 26-28, G.Baritiu Street, RO-400027, Cluj-Napoca, Romania, Tel: +40 264 401 568, Fax: +40 264 591 340 {Anca.Apatean, Simina.Emerich}@com.utcluj.ro

Abstract: In this paper, we mainly concentrate on the recognition module of an object categorization  system. Two types of images, visible and infrared, are investigated in order to improve the objects detection and recognition process. Different types of mother wavelets (Haar, Daubechies, Coiflet, Symlet, Biorthogonal, or fractional causal, anti-causal, generalized and symmetric) are used to extract the wavelet coefficients which will constitute the feature vector. The obtained feature vector then will be fed to a KNN classifier, in order to classify the object in one of the possible object’s classes used in the training step.

Key words: object recognition, wavelets, feature extraction, KNN classifier.

A NOVEL COMBINING TECHNIQUE FOR ADAPTIVE ANTENNA ARRAY

Nicolae CRIŞAN, Ligia Chira CREMENE - Technical University of Cluj-Napoca, 15 Daicoviciu street, Nicolae.Crisan@com.utcluj.ro

Abstract: In the context of new high-data-rate wireless communications, multiple antenna systems are now taken into account early in the design phase. MIMO (Multiple Input Multiple Output) transmission schemes are still based on high redundancy algorithms. This paper proposes a new combining technique, OSC-SSBC (Orthogonal Space Combining – Space-Space Block Coding), and demonstrates that multiple antenna systems can significantly increase channel capacity (close to double), in LOS conditions, by implementing the OSC method at the receiver, on an adaptive antenna array. The block diagram and mathematics are presented and discussed, for a 2Tx-2Rx configuration.

Key words: space diversity, spatial multiplexing, beamforming and beam-steering, OSC – Orthogonal Space Combining, MRC – Maximum Ratio Combining, MIMO – Multiple-Input-Multiple-Output.

Number 1/2008

GERSHGORIN CIRCLES ASSOCIATED TO DOUBLE GRID SECOND ORDER CELLULAR NEURAL NETWORKS

Iolanda ALECSANDRESCU^* Liviu GORAS^*#

*Faculty of Electronics and Telecommunications, “Gh Asachi” Technical University, Iasi, Romania

#Institute for Computer Science, Iasi, Romania

Abstract: The stability of variable parameters double grid second order cell Cellular Neural Networks (CNN’s) linearized in the central linear part of the cell characteristic is investigated by means of Gershgorin’s theorem using spatial domain as well as spatial-frequency domain descriptions. It is shown that the stability margins towards the right-hand side of the complex  plane are identical within both approaches and are larger than expected according to simulations. A conjecture regarding the limits of the characteristic polynomials roots is made and verified through simulations.

Key words: Cellular Neural Networks, eigenvalues, Gershgorin circles.

SPECTRAL CHARACTERIZATION OF TWO FAMILIES OF FIBER OPTIC TRANSMISSION CODES

Nicolae Dumitru ALEXANDRU

“Gh. Asachi” Technical University of Iaşi, Romania, Bd. CAROL I No.11 IAŞI RO-700506, 40-232-213737 Ext.105, Fax: -40-232-217720, nalex@etc.tuiasi.ro

Abstract: Because of the advantages it offers, such as high bandwidth, electromagnetic interference immunity, light weight, etc, the optical fiber is replacing the classic coaxial or twisted pair links. Two families of fiber optic transmission codes are presented and investigated in terms of their spectral properties. The D.C. component carried by these codes affects the power budget of the signal and results in baseline wander that determines an increase of the bit error rate. However, they have the advantage of simple coding and decoding procedures and circuits, which are easy to implement.

Key words: Fiber optic codes, power spectral density, coding factor, autocorrelation function.

A FAST PARALLEL HUFFMAN DECODER FOR FPGA IMPLEMENTATION

Laurentiu ACASANDREI (*), Marius NEAG(**)

(*)Silicon Systems Transylvania SRL, Tel: +40258775181, acasandreilaurentiu@yahoo.com

(**)Technical University of Cluj-Napoca, Tel: +40264401470, Marius.Neag@bel.utcluj.ro

Abstract: The paper presents a novel algorithm and architecture for implementing a Huffman decoder. It starts with an overview of the basics, from the entropy coding and the way the Huffman coding is obtained, to the way a Huffman coder handles data and image components within the Jpeg standard. Then it briefly discusses the decoding procedures proposed by the ISO/IEC 10918-1(1993E) standard; due to their sequential nature, a decoder that simply implements these procedures requires several execution cycles to output one set of decoded data. A new decoding algorithm is then introduced, based on a parallel architecture that allows it to output a set of decoded data per each clock cycle. This approach was validated through actual implementation on an of-the-shelf FPGA; this not only demonstrates the proposed algorithm and architecture but also proves that it can operate at very high frequencies, up to 100MHz. A limitation of this implementation is the relatively large amount of hardware resources it requires.

Key words: Jpeg Standard, Fast Huffman Decoder, Image Acquisition and Processing, Parallel Architecture, FPGA

A VIRTUALLAB APPROACH FOR WIRELESS SENSOR MOTES AND WIRELESS SENSOR NETWORKS

Anghel V. CONŢIU, Adina ŢOPA, Vasile T. DĂDÂRLAT

Technical University of Cluj-Napoca, Str. Ctin Daicoviciu nr. 15, 400027 Cluj-Napoca, Romania E-mail: Vasile.Teodor.Dadarlat@cs.utcluj.ro

Abstract: The VirtualLab project was developed to fill the need of exploring the world of sensors, to make it available to younger students and to increase their interest in studying real sciences like physics, electronics or computer science. It is also a solution for low budget laboratories, making the motes available remotely.

Key words: Mote, Radio, Remote, Sensor, Wireless.

A MULE BASED STRATEGY FOR WIRELESS SENSOR MOTES TO REDUCE POWER CONSUMPTION

Adina ŢOPA, Anghel-Vasile CONŢIU, Vasile Teodor DĂDÂRLAT

Technical University of Cluj-Napoca, Str. C-tin Daicoviciu nr. 15, 400027 Cluj-Napoca, Romania E-mail: Vasile.Teodor.Dadarlat@cs.utcluj.ro

Abstract: The paper presents an application for wireless sensors, which focuses on reducing power consumption of the wireless sensor devices. The project uses a new technology, having a very big potential: motes called Tmote developed at Moteiv. The goal of this application is to investigate the use of the Tmote nodes as mules, to implement a power consuming strategy, so that the sensors deployed through the network will consume as little power as possible.

Key words: wireless sensor motes, power consumption strategy, data collection

UNANTICIPATED DYNAMIC ADAPTATION OF CONTEXT-AWARE SERVICES

Marcel CREMENE* Michel RIVEILL** Christian MARTEL***

*Universitatea Tehnica din Cluj-Napoca, Romania, cremene@com.utcluj.ro** Universite de Nice, Sophia-Antipolis, France, riveill@unice.fr***Universite de Savoie, Chambery, France, christian.martel@univ-savoie.fr

Abstract: Usually, for building context-aware services, the developer must anticipate all the possible contexts elements and their states: user preferences, physical resources, location, etc. This anticipation is required in order to specify service and context-specific adaptation rules, which are necessary for the adaptation management. Because of this, the adaptation rules are not reusable and the service cannot react to unanticipated context changes. In this paper, we propose a platform for dynamic and unanticipated adaptation of component-based services. Our solution is based on a knowledge representation describing the service and the context as a whole system. Based on this service-context representation, the adaptive platform is able to check the adequacy between the service and the context and to search for solutions if necessary. The difference between the classical approach and our approach is that we use only general rules. We have tested our architecture using a prototype that shows that a forum service may be adapted to the user language without using service and context-specific rules.

Key words: Context-aware services, Components, Knowledge representation, Adaptation rules

SEGMENTATION OF COLOR IMAGES USING A COMBINATION BETWEEN MARKOW RANDOM FIELDS AND GRAPHS

Stefan Gherbea (*), Aurel Vlaicu (**)

(*) C.S. Hidroelectrica S.A. – S.H. Cluj, Cluj-Napoca, 1 Taberei st., Tel. 0788202770 Fax:(040) 0264-427797 , e-mail Stefan.Gherbea@hidroelectrica.ro

(**) Technical University of Cluj-Napoca, 15. C.Daicoviciu str.Aurel.Vlaicu@com.utcluj.ro

Abstract: The goal of image segmentation is partitioning the images into homogeneous and interpretable regions. Image interpretation is part of scene understanding. It can be viewed as the process of giving meaning to a 2D image by identifying and labeling significant objects or segments in the image. For example, we may first segment an image and then interpret each segment as being a road, river, building etc.

Key words: MRF, Gibbs distribution, Simulated Annealing, Graph, MST, probability

MODELLING AND SIMULATION OF PHOTOVOLTAIC CELLS

Dorin PETREUŞ, Cristian FĂRCAŞ, Ionuţ CIOCAN

Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology,26-28 G. Bariţiu Street, Cluj-Napoca, Romania, Phone +40-264-401412, E-mail: Dorin.Petreuş@ael.utcluj.ro

Abstract: A photovoltaic cell converts the solar energy into the electrical energy by the photovoltaic effect. Solar cells are widely used in terrestrial and space applications. The photovoltaic cells must be operated at their maximum power point. The maximum power point varies with illumination, temperature, radiation dose and other ageing effects. In this paper, we present four models for a photovoltaic cell. Each method was evaluated and their strengths/weaknesses were identified. Two empirical models was modeled and simulated in Mathcad. The empirical models developed are validating through the comparison of the obtained characteristics with the ones given by the manufacturers of the PV panels.

Key words: photovoltaic (PV) cell, maximum power point (MPP), I-V curves, model.

Number 4/2007

EXTRACTION AND ANALYSIS OF DIURNAL FADE VARIABILITY AND SLOPE FROM SLANT PATH KA BAND RADIOMETER MEASUREMENTS

Ana-Maria Ersilia ANCA

Technical University of Cluj-Napoca, No.15, C. Daicoviciu Street, 400020 Cluj - Napoca, Romania, E-mail: anca_anamaria@yahoo.com

Abstract: The use of Ka-band and higher frequencies in satellite communication systems presents important advantages with respect to lower frequencies. However, the atmospheric-induced attenuation in these bands is substantial, and may lead to frequent outages that will compromise the availability and quality of service. Thus, the use of fade mitigation techniques, in order to compensate for excessive attenuation, will be in many cases unavoidable. The design and implementation of fade mitigation techniques require not only the knowledge of the inter-annual, seasonal and diurnal variability of long-term attenuation statistics, but also of the dynamic characteristics of attenuation, such as fade duration, inter-fade intervals and fade slope.

Keywords: attenuation, fade slope, radiometer, atmospheric remote sensing, cumulative distribution, probability density distribution

DIGITAL FLL DETECTOR FOR SATELLITE BEACON RECEIVERS

Ancuta MOLDOVAN, Tudor PALADE

Technical University of Cluj-Napoca, No.15, C. Daicoviciu Street, 400020 Cluj - Napoca, Romania

No.26-28, Baritiu Street, E-mail: Ancuta.Moldovan@com.utcluj.ro, Tudor.Palade@com.utcluj.ro

Abstract: The paper addresses the design, implementation and test of a beacon receiver coherent detector by using a DSP based software controlled frequency locked loop. The system is built around a user developed card, using fast ADC’s and dedicated signal processing programmable chips, interfaced with a DSP kit. The DSP implements the loop filter by programming in real time a numerically controlled oscillator. The preliminary results are presented and discussed.

Key words: digital receiver, frequency locked loop, loop filter, frequency discriminator, numerically controlled oscillator

FRAME BY FRAME PHONEME CLASSIFICATION USING MLP

József DOMOKOS¹ , Gavril TODEREAN²

¹Sapientia - Hungarian University of Transylvania, Faculty of Technical and Human Sciences, Tîrgu-Mureş , domi@ms.sapientia.ro

²Technical University of Cluj-Napoca, Faculty of Electronics, Telecommunications and Information Technology, toderean@pro3soft.ro

Abstract: In this paper, there are presented some practical experiments for continuous speech frame-by-frame phoneme classification using MLP neural networks. We used to train and test our software application, the TIMIT continuous speech database, and the OASIS Numbers database. In the experiments, we tried to classify all the existing 61 phonemes together, from TIMIT dictionary and the 32 phonemes from OASIS Numbers database. We also used different MLP configurations to compare the achieved results. For classification, we used 13 MFCC coefficients and their first and second order derivatives (delta parameters) extracted from speech signal using a Matlab® based feature extractor software application.

Key words: phoneme classification, MFCC, delta, double delta, MLP, continuous speech recognition, TIMIT, OASIS Numbers;

FPGA IMPLEMENTATION OF THE ADAPTIVE LEAST MEAN SQUARE ALGORITHM

Irina DORNEAN, Marina TOPA, Botond Sandor KIREI, Erwin SZOPOS

Bases of Electronics Department, Technical University of Cluj-Napoca, Romania

26-28 G. Baritiu str., Cluj-Napoca, Romania, Tel: +40264401470; Fax: +40264591340 Irina.Dornean@bel.utcluj.ro

Abstract: The paper presents a digital implementation of the adaptive Least Mean Square (LMS) algorithm. The algorithm was chosen due to its robustness and simplicity. The envisaged application is the identification of an unknown system. The implementation was done in two steps: first the algorithm is checked using a model in Matlab Simulink and then the Verilog code is written and tested. The results prove the theory that while the error converges to a minimum value, the output and the desired signal are comparable in shape and values. The application enables further developments and researches, creating an opportunity for automation and prediction.

Keywords: LMS, Simulink, FPGA

ESTIMATION RELIABILITY OF MICROARRAY DATA USING HIGH ORDER STATISTICS

R. MALUTAN^1,2, P. GOMEZ VILDA², M. BORDA¹, F. DIAZ²

¹Universitatea Technica Cluj Napoca, Geroge Baritiu nº 26-28, 400027, Cluj Napoca, Romania

²Universidad Politecnica de Madrid, Campus de Montegancedo, s/n, 28660, Boadilla del Monte, Spain

raul.malutan@com.utcluj.ro

Abstract: Microarray technology posses some interesting problems under the point of view of Genomic Signal Processing, as test probes express themselves in rather different patterns, not showing proportional expression levels for most of the segment pairs, as it would be expected. These cases may be influenced by unexpected hybridization dynamics, and are worth of being studied with a double objective: gain insight into hybridization dynamics in microarrays, and to improve microarray production and processing. Two methods are proposed: modeling the dynamics of expression dynamics and isolating gene expressions showing unexpected behavior to proceed in their further classification and study.

Key words: microarray, hybridization, orthogonality, reliability.

ANALYSIS OF HIGH POWER MEMS VARACTORS FOR RF AND MICROWAVES INTEGRATED CIRCUITS

Nicolae CRIŞAN, Ligia CREMENE

Technical University of Cluj-Napoca, 26-28, Baritiu street, phone: 40264401914, crisan@com.utcluj.ro, ligia.cremene@com.utcluj.ro

Abstract: This paper focuses on designing, modelling and analysis of a MEMS capacitive cantilever switch using EM-CAD simulators. Capacitive MEMS have many applications in adaptive filters and smart antennas. EM-CAD simulators can help determine the S dispersion parameters for MEMS devices. Modelling MEMS switches by S-parameters, enables the usage of MEMS directly into electronic-circuit-oriented simulators (non - EM simulators), based on S-parameter measurements. The proposed method enables emulation of MEMS devices in any

circuit-oriented CAD analyzer that uses quadripol models instead of MEMS component models. After importing the S-parameter files from one simulator to the other, the analysis of the most complex circuits (MEMS together with classic electronic components) becomes possible. This analysis remains valid for the frequency band in which the MEMS device has been analyzed. Adaptability of the resulted circuit is the main achievement of the process.

Key words: MEMS – Micro-Electro-Mechanical Systems, EMF – Electromagnetic Field, FEM – Finite Element Method

MODEL-BASED DESIGN AND FPGA IMPLEMENTATION OF CONTROL SYSTEMS

Gabriel OLTEAN*, Botond Sandor KIREI*, Irina DORNEAN*, Ioana OLTEAN**

* Bases of Electronics Department, Technical University of Cluj-Napoca 26-28 G. Baritiu str., Cluj-Napoca, Romania, Tel: +40264401470; Fax: +40264591340; Gabriel.Oltean@bel.utcluj.ro

** Grup Şcolar Electrotehnic „Edmond Nicolau”, Cluj-Napoca

Abstract : FPGAs have become a very promising solution for the realization of digital control systems. In this paper an efficient model-based design methodology for FPGA implementation of a control system is presented. The Matlab/Simulink environment is used here for modeling, simulation and tuning a temperature control system based on PID controller. Verilog HDL language describes the digital control system to be implemented on a FPGA development board. Results prove that the proposed methodology shortens the design cycle, because the whole system can be fully simulated and adjusted before its final design and implementation.

Keywords: PID controller, Verilog, FPGA

METHODS FOR SINGING VOICE IDENTIFICATION USING ENERGY COEFFICIENTS AS FEATURES

Annamaria MESAROS, Simina MOLDOVAN

Technical University of Cluj-Napoca, ROMANIA, G. Baritiu 26, RO-40027

Abstract: The problem of singer identification becomes a specific point of interest in themusical signal analysis field. This paper describes two energy representations of the voice signal and tests their efficiency in singing voice identification. The first set of energy features consists in the Mel-scale energies of 14 frequency bands, covering the whole frequency spectrum of the signal. The second energy representation is obtained by wavelet decomposition of the voice signal. The wavelet and scaling filters for the decomposition are derived from fractional B-spline functions. The wavelet decomposition is done hierarchically, into 14 bands, with octave-band filters, taking into account the specific frequencies of the formants. Both energy representations are tested for singing voice identification on the training set and on unknown data.

Number 3/2007

IP CAMERA SURVEILLANCE SYSTEM WITH AUTOMATED RECORDING

Mihai ROMANCA, Iuliu SZEKELY, Sorin COCORADA, Iulia GRAMA

Transilvania University of Brasov, 29 Eroilor, Romania, Tel./Fax: 0268 478705, romanca@unitbv.ro

Abstract: In this article we present the results obtained at the installation and configuration of a video surveillance system for controlling the access in a building. The system uses an IP video camera (NC 1200) connected at a building LAN. The IP camera used allows the visualisation of images by Internet connection but, for recording images, admits only manual command from the human operator that supervises the entrance. The camera has embedded the function of sending an instant picture by e-mail or on a FTP server if the system disposes any of these two facilities. For this reason we created an application that automatically starts recording images, when a motion is detected in the supervised space. The application has the advantage of saving the image frames directly on the monitoring computer and eliminates the constraint to have an online server (FTP or e-mail) and also that permits no modification of the camera firmware, being installed only on the monitoring computer. We report here some preliminary results, with only one surveillance camera, but the system will be extended to several ones. We illustrate the advantage of using a cheap IP camera combined with the original application for entrance surveillance.

Keywords: intelligent building, IP camera surveillance, real-time, intelligent visual surveillance

MONITORING LOW-RATE WIRELESS PERSONAL AREA NETWORKS USING SNMP

Cristian M. VANCEA, Virgil DOBROTA

Technical University of Cluj Napoca, Communications Department,

George Baritiu 26-28, 400027 Cluj-Napoca, Romania, Tel: 40-264-401264, Fax: 40-264-597083

E-mail: {Mihai.Vancea, Virgil.Dobrota}@com.utcluj.ro

Abstract: This paper presents a practical approach for enabling Simple Network Management Protocol for devices using IEEE 802.15.4 standard. This work is focused on providing a system that can be used for management of low-rate wireless personal area network equipment.

Key words: embedded systems, network management, Personal Area Network, SNMP

EVALUATION OF QoS PARAMETERS FOR IPTV

Ancuta Sanda BUZILA, Gabriel LAZAR, Tudor BLAGA, Virgil DOBROTA

Technical University of Cluj Napoca, Communications Department

George Baritiu 26-28, 400027 Cluj-Napoca, Romania, Tel: 40-264-401264, Fax: 40-264-597083

E-mails: anca_buzila@yahoo.com, {gabriel.lazar,tudor.blaga,virgil.dobrota}@com.utcluj.ro

Abstract: This paper describes in details the IPTV system, together with its implementation methods and the involved protocols. Furthermore, measurement methods of QoS parameters are analyzed, discussing a practical application for evaluation of IPTV traffic parameters. The following components are taken into account: bandwidth, one-way delay, inter-packet jitter, packet loss, number of duplicated packets, out-of-order packets and corrupted packets. In addition, two QoS parameters of importance for VoD are measured: START delay and PAUSE/RESUME delays. The implementation of this program opens the possibility of measuring the transmission parameters that influence QoS, with the purpose of offering a way to evaluate and control the services delivered by open-standards IPTV providers.

Key words: IPTV, measurements, QoS, XML

MULTICAST vs. UNICAST TRANSMISSIONS FOR WIRELESS IP CAMERA SURVEILLANCE SYSTEMS

Sorin COCORADĂ

University Transilvania of Brasov, 29 Eroilor, Romania, Tel./Fax: 0268 478705, sorin.cocorada@unitbv.ro

Abstract: This paper presents the results obtained at the installation, configuration and optimization of a wireless IP video surveillance system. Because of the cheap and insufficient hardware resources, the WVC54GC camera permits only a very limited number of HTTP over TCP unicast streams. Multicasting at the data link layer is a more efficient way for sending data to a group of receivers. Some possible solutions for overcoming the above limitation are presented using additional hardware and software. An analysis of their performance is presented using network monitoring tools, together with a discussion. Results have shown that there are several alternatives to efficiently provide a video stream to the end users.

Key words: wireless IP camera, unicast, multicast, IEEE 802.11

ASPECTS OF ENTERTAINMENT DISTRIBUTION IN AN INTELLIGENT HOME ENVIRONMENT

Radu ARSINTE

Technical University Cluj-Napoca, Tel: +40-264-595699, Str. Baritiu 26-28, Radu.Arsinte@com.utcluj.ro

Abstract: The paper presents an implementation and tests of a simple home entertainment distribution architecture (server + multiple clients) implemented using two conventional cabling architectures: CATV coaxial cable and conventional Ethernet. This architecture is created taking into account the “Home gateway” concept present in most attempts to solve the problem of the “Intelligent home”. A short presentation of the experimental is given with an investigation of the main performances obtained using this architecture. The experiments revealed that this simple solution makes possible to have entertainment and data services with performances close to traditional data services in a cost-effective architecture.

Key words: Home network, Streaming, Ethernet, Home Gateway

SELECTIVE DISTANCE DETECTING AND LOCATING SYSTEM

Gheorghe MORARIU, Marian ALEXANDRU, Adrian MAILAT

Transilvania University, Brasov, Romania 1 – 3 Politehnicii str., Phone/Fax +40268478705

gmorariu@vega.unitbv.ro, alexandrum@vega.unitbv.ro

Abstract: The goal of this paper is to present a selective distance detecting and locating device capable of determining the position of a fixed or moving target situated inside of a monitored area of given width placed at a certain distance from the transmitter/receiver antenna. The locator unit performs both the location and ranging function and simultaneously identifies the target status (mobile/immobile) and nature (biological/non-biological).

Key words: locating system, ultrasound, remote control

REMOTE TEMPERATURE RECORDING USING BLUETOOTH TECHNOLOGY

Marian ALEXANDRU, Mihai ROMANCA, Petre OGRUTAN

Transilvania University, 1 – 3 Politehnicii Str Brasov, Romania Tel/Fax +40268478705,

alexandrum@vega.unitbv.ro; romanca@unitbv.ro

Abstract: The past decade has seen significant advancement in the field of consumer electronics. Various “intelligent” appliances such as cellular phones, air-conditioners, home security devices, indoors localization, etc. are set to realize the concept of a smart home or building. A wireless home network that does not incur additional costs of wiring would be desirable. Bluetooth technology, which has emerged in late 1990s, is an ideal solution for this purpose. This paper describes an application of Bluetooth technology in home or industrial automation networking environment. It proposes a network, which contains a remote temperature sensor connected to a microcontroller, which controls, also, the Bluetooth module transmission and a computer having internet access, connected via USB to a Bluetooth module. In this way, the sensor data are collected and processed by the microcontroller, transmitted via Bluetooth to a fixed or mobile computer and published on a web page in order to be accessible for a remote user.

Key words: Bluetooth, smart building, temperature recording

A SYSTEM FOR BUILDING AUTOMATION SUPPLIED FROM A SOLAR PANEL

Dorin PETREUS, Niculaie PALAGHITA, Zoltan JUHOS

Faculty of Electronics, Telecommunications and Information Technology Technical University of Cluj-Napoca, Cluj-Napoca, Romania

Dorin.Petreus@ael.utcluj.ro

Abstract: This paper presents a solar power based scalable automation module designed for sensor/actuator interfacing and communication in intelligent buildings. The system was designed based on a number of modules which are connected together. This gives to the system more flexibility. A complete system (hardware and software) was implemented.

Key words- intelligent building; photovoltaic panel; battery powered; FSK modem.

OPEN SYSTEMS IN BUILDING MANAGEMET: LONWORKS vs. BACNET

Tudor SAMUILĂ, Bogdan ORZA, Aurel VLAICU

Technical University of Cluj-Napoca 15 C. Daicoviciu Street, 400020 Cluj-Napoca, Romania

Email: Tudor.Samuila@com.utcluj.ro

Abstract: The LonWorks and BACnet technologies gained widespread application and are the best choice as the leading open system

technologies. They have distinguished above the rest with a flexible architecture focused on meeting the growing requirements of building automation. Despite their achievements, LonWorks and BACnet have very little in common. They have different origins, different network architectures, different hardware and software requirements, and each has a core group of users. A comparison relevant to building automation should focus on market presence and solution approach. A market presence comparison is an indication of user and manufacturer acceptance through the number of user organizations and product availability. A solution approach comparison illustrates the differences and implementation issues to consider when evaluating a solution.

Key words: LonWorks, BACNet, building management, open systems, comparison

EMBEDDED WEB-SERVERS FOR REMOTE CONTROL IN DOMOTICS

Florin SANDU ⁽¹⁾ Daniel IOLU ⁽²⁾

⁽¹⁾ "Transilvania" University - Bd Eroilor nr. 29A, Brasov–Romania, phone/fax +40268478705, sandu@ unitbv.ro

⁽²⁾ Siemens Program and System Engineering, Bd Kogalniceanu nr.21, Brasov-Romania,

phone +40268409019, fax +40268409221, daniel.iolu@siemens.com

Abstract: In the frame of the CEEX –60- II03 INFOSOC contract 60- II03 – COMODICI, "Remote Control and Monitoring of Intelligent Buildings", the authors accomplished device enablement and device networking by embedded web-servers with serial tunneling. Their focus is on domotic devices but other appliances (remote testing in communications and remote health monitoring) are also mentioned

Key words: Domotics, device servers, serial tunneling, remote switching

A MUTISENSOR GPRS-BASED SECURITY SYSTEM FOR INTELLIGENT BUILDING

Petre OGRUTAN, Mihai ROMANCA, Csaba-Zoltán KERTÉSZ

TRANSILVANIA University of BRASOV, Department of Electronics and Computers Postavarului 3 ,tel/fax +40-268478705

ogrutan@vega.unitbv.ro, romanca@vega.unitbv.ro csaba.kertesz@vega.unitbv.ro

Abstract: In this work we present a security system collecting data from detectors for carbon monoxide, flammable gas, smoke, radon and from an access control system based on fingerprint recognition, and sending warnings through a GPRS link. The warnings are sent over the Internet is the concentration of toxic gases or smoke increases over a certain level or an unauthorized access is repeated several times. The transmitted data contains information about the type of danger situation and the location of the system determined by a GPS receiver.

Key words: Intelligent building, GPRS transmission, embedded system, Radon detector

REDUNDANCY FOR THE SECURITY SYSTEMS IN A SMART BUILDING USING INTERNET (WEB) AND GSM (SMS) TECHNOLOGIES

Mugur MOCOFAN, Radu VASIU, Mircea ABUCEANU

Politehnica University of Timisoara, Electronics and Telecommunications Faculty

Bd. Vasile Parvan, Nr. 2, Timisoara, Tel: +40 256 403319, Fax: + 40 256 403300
mugur@cm.upt.ro radu@cm.upt.ro mircea@multimediadesign.ro

Abstract: Security is a problem at multiple levels and for this reason we create an isolated, parallel network to control the smart buildings using the SMS technologies.

Key words: Smart buildings, SMS server, GSM.

SCENERIES CONTROL FOR SMART BUILDINGS USING RELATIONAL DATABASES

Muguras MOCOFAN, Radu VASIU, Iulia RADU

Politehnica University of Timisoara, Electronics and Telecommunications Faculty

Bd. Vasile Parvan, Nr. 2, Timisoara, Tel: +40 256 403319, Fax: + 40 256 403300
mugur@cm.upt.ro radu@cm.upt.ro jyar_33@yahoo.com

Abstract: The modernization of the buildings with the most performing technologies starts to grow more and more, and like a consequence, many people appeal to this technique, to leave in a more safety house, and why not, more comfortable. We propose a relational database for controlling the sceneries of smart buildings. The sceneries include all the settings for the house (lights, temperature and music) in a special case, for example “romantic dinner”.

Key words: Databases, Smart buildings, Sceneries.

Number 2/2007

STUDY OF A HYBRID - ANALOG TV AND ETHERNET- HOME DATA LINK USING A COAXIAL CABLE

Radu ARSINTE

Technical University Cluj-Napoca , Tel: +40-264-595699, Str. Baritiu 26-28, Radu.Arsinte@com.utcluj.ro

Abstract: The paper presents an implementation and compatibility tests of a simple home network implemented in a non-conventional manner using a CATV coaxial cable. Reusing the cable, normally designated to supply RF modulated TV signals from cable TV networks, makes possible to add data services as well. A short presentation of the technology is given with an investigation of the main performances obtained using this technique. The measurements revealed that this simple solution makes possible to have both TV and data services with performances close to traditional home data services: cable modems or ADSL, with minimal investments. This technology keeps also open the possibility for future improvements of the network: DVB-C or Data via Cable Modems.

Key words: Hybrid TV Network, Home network, Coaxial Cable, Ethernet

WEB ANALYTICS: A SHORT OVERVIEW (I)

Petre G. POP

Technical University, Cluj-Napoca, Comm. Dept., E-mail:Petre.Pop@com.utcluj.ro

Abstract: Web analytics is the measurement of the behavior of visitors to a website to assure its performance, gain insights into the needs and wants of users' of the site, and identify areas that can be optimized. In a commercial context, it especially refers to the measurement of which aspects of the website work towards the business objectives; for example, which landing pages encourage people to make a purchase. Many different vendors provide web analytics software and services. The purpose of this paper is to present the technical aspects of web analytical methods, then to survey some of these techniques and how they are currently being applied.

Key words: Logfile, Page Tracking, Cookies, Tags, Web Metrics.

5GHz INDOOR AND OUTDOOR MEASUREMENTS

Ligia CREMENE, Nicolae CRIŞAN, Tudor PALADE

Technical University of Cluj-Napoca,Baritiu 25-26 , Cluj-Napoca, phone: 40264401404, Ligia.Cremene@com.utcluj.ro

Abstract – We have tested an 802.11a PTP single-user link for both indoor and outdoor scenarios. We have tried to establish the optimum positioning of the 802.11a WLAN access point at one floor of an office building, observing the influence of people moving around, crowding certain rooms at certain moments of the day, and of course the influence of structure materials. Significant improvements in SNR levels have been noticed when changing the antenna type and modifying the position of the access point. We were thus able to extend the wireless access to other buildings, for LoS (Line-of-Sight) scenarios, and scarce out-of-leaf vegetation.

Key words: 802.11a WLAN, 5GHz measurements, optimum positioning, indoor and outdoor testing.

USING ADVANCED EM-CAD PROGRAMS TO ESTIMATE SPECIFIC ABSORBTION RATE OF THE HUMAN BODY

Nicolae CRIŞAN

Technical University of Cluj-Napoca,Baritiu 25-26, Cluj-Napoca , phone: 40264401404, crisan@com.utcluj.ro

Abstract: In order to prevent adverse health effect of the electromagnetic fields (EM) this article proposes one of the low cost and effective method to simulate and diminish the SAR (Specific Absorption Rate) by means of CAD modelers, taking into account the RF power transmitted and the antenna for a radio device before its hardware implementation. Especially, due to the rapid growth of mobile terminals in the last few years there is a growing concern about possible adverse effect of radio radiation from such devices that usually work closely by the human brain. There are already some reports describing powerful methods for measuring partial body absorption rate associated with the mobile phone terminals. The standardization process is near completed by the Telecommunications Technology Council which has taken this great responsibility.

Key words: SAR –Specific absorption rate, EM – Electromagnetic, DUT – Device under test, Voxel (3D-point)

ACOUSTIC DATABASE FOR ROMANIAN TTS SYNTHESIS. DESIGN AND REALISATION RESULTS I.

Árpád Zsolt BODÓ, Ovidiu BUZA, Prof. Gavril TODEREAN

Technical University of Cluj-Napoca, Faculty of Electronics and Telecommunications zsolt.bodo@gmail.com; ovidiu.buza@cs.utcluj.ro; toderean@pro3soft.ro

Abstract: The current paper presents the general design and partial realisation of a diphone database for Romanian concatenative text-to-speech synthesis. The design of a database and the planning of its realisation are presented in detail. The basis for the decisions, the aspects that need to be considered during the design, the rules elaborated – for e.g. the constructions of senseless words – are focused as well. The steps of realisation are shown theoretically and by practical hints too. However the database is not completely finished, the accomplished experiences during its development are shared by this paper.

Key words: Speech synthesis, Diphone, Database, Intrinsic prosodic parameters, Intonation, Pitchmarking, Senseless words, DB element processing, Monotone, Design checklist, Carrier text, Recording, Extraction, Automation.

REALISATION RESULTS OF A SPEECH SYNTHESIS DEVELOPMENT ENVIRONMENT

Árpád Zsolt BODÓ, Ovidiu BUZA, Prof. Gavril TODEREAN

Technical University of Cluj-Napoca, Faculty of Electronics and Telecommunications zsolt.bodo@gmail.com; ovidiu.buza@cs.utcluj.ro; toderean@pro3soft.ro;

Abstract: The development of a text-to-speech synthesis system is a complex process that is elaborated in several fields like: speech signal processing, prosody prediction and generation, synthesis techniques/engines, database construction, etc. Each of these domains, during their research and development phase, requires a toolset. The need for this is multiple: some times intermediate results are required to be visualised, some times external (manual) interaction to the system helps the researcher’s work. The current paper shows a collection of these kinds of tools, which finally form a development environment. This framework is built around a general voice signal editor freeware software, but the TTS specific tools and graphical user interfaces are the result of an own development. The main fields detailed below are: the PSOLA synthesis technique development environment, the diphone database processing environment, the speech signal processing toolset, some prosody modification related issues and intonation curve computation algorithm.

Key words: TTS development environment/framework, Toolset , Synthesis technique, Waveform, Intonation curve, Pithmarking, Prosody manipulation, Pitch descriptor, Prosodic descriptor, Temporal structure, Speech signal processing, Waveform editing.

FPGA IMPLEMENTATION OF PWM MODULES

Bogdan Stefan MEREUTA¹, Mihai LUCANU², Tecla Castelia GORAS³

¹ Technical University Iasi, Faculty of Electronics and Telecommunications, Blvd. Copou No. 11, 6600, Iasi, Romania, +40740368957, bmereuta@etc.tuiasi.ro

^{2, 3} Technical University Iasi, Faculty of Electronics and Telecommunications, Blvd. Copou No. 11, 6600, Iasi, Romania

Abstract: This paper presents a new method of PWM implementation with FPGA. Most FPGAs contain DSP blocks which minimize the die’s logic area and improve the PWM speed. The correct operation of the PWM module was verified by simulation.

ALGORITHM FOR DETECTING INTERSECTION POINTS BASED ON SWEEP LINE, WHEN THE SEARCH SPACE CONTAINS BOTH SEGMENTS AND ARCS

Aurel VLAICU¹, Stefan GHERBEA²

¹Technical University of Cluj-Napoca, e-mail Aurel.Vlaicu@com.utcluj.ro

²C.S. Hidroelectrica S.A.–S.H. Cluj, Cluj-Napoca, Stefan.Gherbea@hidroelectrica.ro

Abstract: The problem of the intersection point bet,een segments is very important in computer graphics. When displaying objects on 2 dimensional devices, intersecting sections should be eliminated, in order to enhance realism or to optimize the visualization. For example when a vector image is large and we need to view only a little portion from it, thehidden portion of the image should be eliminated. In 2002 Susa Hert – Kurt Mehlhorn developed a generalized algorithm for detecting intersection between conics arcs, based on sweep line. We propose a particularized version of the algorithm, for segments and arcs and a pseudo – code will be proposed. Based on this algorithm, an optimization for polyline outline structures was realized.

Key words: Arcs, segments, sweep line ,graphics algorithm, C++

Number 1/2007

ON THE PROPERTIES OF SOME PARTICULAR DATA SEQUENCES USED IN UWB-PPM SYSTEMS

Tudor DURNEA

''Gh. Asachi" Technical University of lasi. Faculty of Electronics and Telecommunications. Carol I avenue, nr. 11. RO-700506. e-rnail: dtudor@etc.tuiasi.ro

Abstract: The initial UWB systems based on impulses tried to implement PPM - type modulations in order to achieve maximum benefit from available large bandwidth. The UWB-PPM systems were noted to generate large interference to narrowband systems operating in the same band. The cause of this interference is the spectrum generated by the PPM modulation as it is known to concentrate high amounts of energy at discrete frequencies. There is no complete formula for the spectrum of the PPM signals for any symbol probabilities and any number of modulation bits. We present some particular properties of PPM symbols as we need them to evaluate their spectrum in order to successfully implement UWB-PPM systems.

Key words: UWB, PPM, Discrete Spectrum Components, Diff.

CONTENT BASED MEDICAL IMAGE RETRIEVAL USING ORACLE INTERMEDIA

Ligia CHIOREAN, Aron SIPOS, Mircea-Florin VAIDA

Technical University of Cluj-Napoca, Baritiu Street no. 26, 400027, Cluj-Napoca, Romania E-mails: Chiorean.Ligia@com.utcluj.ro, siposa@yahoo.com, Mircea.Vaida@com.utcluj.ro

Abstract- Oracle database offers the facility of searching and retrieving content-based multimedia objects such as audio, video, image or electronic documents, after a signature. There are different methods for signature generation for each specified object. The content-based image retrieval (CBIR) in Oracle may be realized on a linear search and comparison of the signatures using different parameters, or by image indexing, which are methods available in Enterprise Edition of Oracle Database. medBase is a dedicated application that uses Oracle interMedia for CBIR. Oracle has the possibility to integrate Java classes that offers more flexibility to a programmer in generating and processing the database objects. The chosen technology implemented in this paper is based on JSP, which is capable of realizing distributed web applications, and it can directly access the Java classes and methods offered by Oracle. The dedicated application was tested using different types of medical images stored in an Oracle database.

Key words: Content-based retrieving, multimedia database, Oracle interMedia

AN INVESTIGATION OF ISI AND ICI PROPERTIES FOR A FAMILY OF OFFSET RAMP IMPROVED NYQUIST FILTERS

Ligia Alexandra ONOFREI(*), Nicolae Dumitru ALEXANDRU(**)

(*)"Stefan eel Mare" University of Suceava, str. Universitafii no. 13, RO-720225, Suceava, onofreial@eed.usv.ro,

(**)"Gh. Asachi" Technical University oflasi,, Bd. Carol 1, no 11, RO-700506 Iasi, nalex@etc.tuiasi.ro

Abstract: We proposed a family of new Nyquist pulses based on an offset ramp method and investigated its performances in terms of ISI error probability. These pulses proposed and studied are employed for OFDM use, to reduce the sensitivity of OFDM systems to frequency offset. The results presented in this paper equal or outperform the recently found pulses in terms of intercarrier interference (ICI) power.

Key words: intersimbol-interference (ISI), intercarrier-interference (ICI), frequency offset

EXPERIMENTAL PERFORMANCE DETERMINATION OF END SYSTEM MULTICAST AND HYPERCAST

Tudor Mihai BLAGA, Gabor HEGEDUS, Virgil DOBROTA

Technical University ofCluj-Napoca, Communication Department, G. Baritiu Street, no. 26-28, tel. +40-264-401816,

Email: tudor.blaga; virgil.dobrota@com.utcluj.ro;hegegabo@yahoo.com

Abstract: End System Multicast (ESM) and HyperCast are overlay multicast proposals. Data forwarding is performed along paths from the overlay topology, which is built over the physical network topology and includes only end nodes. In this paper we study the performance and the efficiency of these two proposals. We use the following metrics: control overhead, time to first packet, stress, resource usage and stretch to determine the performance. Parts of the results are obtained using LOTOS (LOgical TOpology Simulator) from HyperCast.

Key words: Multicast, overlay.

SENSORS FOR OBSTACLE DETECTION IN TRAFFIC SCENE SITUATION

Anca DISCANT¹, Alexandrina ROGOZAN², Corneliu RUSU¹, Abdelaziz BENSRHAIR²

1. Technical University of Cluj-Napoca, Romania, Baritiu str, 26-28, 400027 Cluj-Napoca, Romania, ancad@bel.utcluj.ro

2. National Institute of Applied Sciences, Rouen, France

Abstract: The main sensors types used for the obstacle detection field are presented. They can be classified as proprioceptive or exteroceptive sensors, but also in two important categories: passive sensors (the cameras using visible spectrum-CCD and infrared cameras) and active sensors (radar, laserscanner, sonar). Generally, more camera types or configurations can be used in the obstacle detection domain: static or moving camera, monocular or stereo images, sequences or simple images, visible or infrared. Although, some segmentation methods used here are: stereo-based segmentation (the depth information of the objects), video-based segmentation (the motion information) and infrared-based segmentation (thermal information of objects).

Key words: exteroceptive and proprioceptive sensors, radar, laser, sonar, infrared camera.

DESIGN AND PERFORMANCE ANALYSIS OF REVERBERATION ALGORITHMS

Norbert TOMA, Marina Dana TOPA, Erwin SZOPOS

Faculty of Electronics, Telecommunications and Information Technology, Technical University ofCluj Napoca,

Str. G. Baritiu nr. 26-28, 400027, Cluj Napoca, Romania, phone: +40 264 401 243; fax: +40-264 591689

email: Norbert. Toma@bel.utcluj.ro, Marina. Topa@bel.utclaj.ro, Erwin. Szopos@bel. utcluj. ro

Abstract: The paper deals with the most important algorithms that simulate natural room reverberation. It starts with the description of the parameters used to evaluate the acoustics of concert halls followed by the presentation of a number of artificial reverberators. The structure, their design and a comparative analysis are given.

Key words: Early reverberation, late reverberation, comb filter, all-pass filter, acoustic parameters.