Institut de Génie Electrique et d'Electronique
Permanent URI for this communityhttps://dspace.univ-boumerdes.dz/handle/123456789/2344
Browse
237 results
Search Results
Item Text dependent speaker recognition using HTK(2016) Bouguerra, Sara; Sahmi, Nacera; Dahimene, Abdelhakim (supervisor)The objective of the project is to design a text dependent speaker recognition system using HTK, the HTK toolkit is used essentially for speech recognition, after understanding the theory of Hidden Markov Models, and the different processes HTK uses during recognition, the principle of Log likelihood probability, and the language Model are linked together to identify the speaker of an intended sentence chosen by the user and trained by the speaker wanted to be recognized. After several experiments, many threshold options were used, till arrived to the kinds of threshold playing a major role in speaker identification, a method for choosing thresholds has been formulated and it was applied for 10 speakers. The method for speaker identification worked successfully for all speakers, and the goal of recognizing a sentence that belongs to a particular speaker has been realized.Item Microstrip band-pass filters for dual band applications, design and implementation(2016) Mermoul, Ali; Hocine, Kenza; Challal, M. (Supervisor)In this work, three new different microstrip dual-band bandpass filters (BPFs) operating at (2.1 GHz, 5.1 GHz), (1.6 GHz, 5.15 GHz) and (2.45 GHz, 5.5 GHz) for 3G, Wi-Fi, WLAN and satellite communications are investigated and realized. The three filters are obtained from modified second order hairpin filter. The proposed filters present satisfactory performances and meet the requirements of modern wireless communication systems. The substrate material used for all designs is FR4 with a thickness of 1.62 mm, a relative dielectric constant of 4.3 and a dielectric tangent loss of 0.0017.Item FIR filter design using optimization algorithms(2016) Zebara, Oussama; Bouhamoum, Youcef Zakaria; Daamouche, A. (Supervisor)FIR Filters are used, nowadays, in different applications because of their phase linearity as well as their intrinsic stability. But for designing near optimal FIR Filters complex mathematical calculations are required, especially for high order FIR Filters. In this report we try to rely on two evolutionary algorithms, namely Particle Swarm Optimization (PSO) and Cuckoo Search Algorithm (CS), to solve this problem using simple procedures. Given the desired specifications to be met by the filter, simple procedures are used to optimize the coefficients approximating the ideal specifications and each algorithm uses its own rules to explore the search space efficiently. The results obtained are compared to PM algorithm in order to assess the performance of the proposed procedure in terms of the magnitude of ripples, transition band and convergence rate.Item Study and simulation of the link adaptation technology in the LTE standard(2016) Meziane, Mokrane; Allilat, Fetta; Zitouni, Abdelkader (Supervisor)The Long Term Evolution first introduced by the link adaptation system as a massive step forward in the control of the network resources. It is mainly based on measuring instantaneous Signal to Interference and Noise Ratio (SINR) which is used to investigated the influence of several parameters on the link adaptation error characteristics. Adaptive Modulation and Coding (AMC) is used to increase the network capacity or downlink data rates. Spatial multiplexing techniques for MIMO antenna configurations are also considered. This report has outlined the various scheduling techniques used in the link adaptation process to increase the throughput of the network. Three major channel state report measurements are used to perform the adaptations which are: CQI (channel quality indicator), PMI (Precoding Matrix Indicator), and RI (Rank Indicator) measurements. An analysis and a comparison of these scheduling algorithms were done through simulations executed on MATLAB software. The impact of the scheduling schemes was examined on the throughput (Average Data Rate) and the Bit Error Rate (BER), and it was compared to the case where no adaptation is done. The fairness of the results given by each scheduling scheme was investigated. The link adaptation improved the transmission robustness through minimizing the error and increased the spectral efficiency to produce a satisfactory average data.Item Digital Filters for reducing the effects of dispersion in optical communication(2016) Grainat, Youcef; Mebarki, Yaser; Zitouni, Abdelkader (Supervisor)The aim of this project is to use digital filters to reduce the effects of dispersion in modern WDM optical communication systems at two transmission speeds, 2.5 and 5 Gbps. For this purpose two types of filters [Gaussian Filter (GF) and Cosine Roll Off Filter (CROF)] have been tested by simulation using OptiSystem and Matlab in single mode fiber links at the operating wavelength of 1550 nm. From the obtained results a comparative study with respect to the conventional Dispersion Compensating Fibers (DCF) has been carried out.Item Study of plus slotted microstrip antenna for wireless applications(2016) Bencharef, Fatma Zohra; Haouchine, Leila; Azrar, Arab (Supervisor)This project deals with the design and simulation of a New Microstrip Patch Antenna using IE3D software. The designed antenna was given a name of "Plus Antenna for Wireless Applications". A rectangular antenna with an operating frequency of 2.4 GHz has been studied at the beginning. Later, the rectangular shape is modified to Plus Shaped Antenna and its study has shown two frequencies. The analysis is then extended to Plus Shaped Antenna loaded by two slots. The obtained results show that the antenna can operate at to 1.72 GHz that can be used in third generation 3G application and 2.98GHz useful in WiMax applications. In addition, the reduction in the operating frequency from 2.4 GHz to 1.72 GHz has allowed size reduction of the structure of about 52%.Item Fusion of voice and face using artificial neural networks at feature level(2016) El Affifi, Omar Badis; Boushaba, Saddek; Cherifi, Dalila (Supervisor)Lately, human recognition and identification has acquired much more attention than it had before, due to the fact that computer science nowadays is offering lots of alternatives to solve this problem, aiming to achieve the best security levels. One way is to fuse different modalities as face, voice, fingerprint and other biometric identifiers. The topics of computer vision and machine learning have recently become the state-of-the-art techniques when it comes to solving problems that involve huge amounts of data. One emerging concept is Artificial Neural networks. In this work, we aim to use both human face and voice to design a multibiometric recognition system, the fusion is done at the feature level with three different schemes namely, concatenation of pre-normalized features, merging normalized features and multiplication of features extracted from faces and voices. The classification is performed by the means of an Artificial Neural Network. The system performances are to be assessed and compared with the K-nearest-neighbor classifier as well as recent studies done on the subject. An analysis of the results is carried out on the basis Recognition Rates and Equal Error Rates.Item Low density parity check coding(2016) Brahimi, Mohamed Amine; Younsi, Bilel; Dahimene, Abdelhakim (Supervisor)Since their rediscovery by Mackay Neal in 1993 [11], LDPC codes have been an integral part of the coding field, and by now, they have been in many standards and are suggested for others. In terms of performance, many practical results have shown that those codes are actually very performant at rates that are within a fraction of capacity. In this Master Project report, the general theory of LDPC codes, including construction, encoding and decoding procedures, is reviewed in the first chapters. In the simulation chapter, we consider the evaluation of error performance of short to medium length LDPC codes with BPSK modulation through the binary AGWN channel.Item Hidden biometrics for identification using ECG and EEG signals(2016) Adjerid, Chaouki; Boukerma, Billal; Cherifi, .Dalila (supervisor)Security concerns increase as the technology for falsification advances and biometrics provides airtight security by identifying an individual based on the physiological and/or behavioral characteristics. Physiological hidden biometrics represented by ECG and EEG biomedical signals are highly confidential, sensitive, and hard to steal and replicate, and also hold great promise to provide a more secure biometric approach for user identification and authentication. This work proposes the human heartbeat as a characteristic to be used for identity recognition. An ECG-based biometric identification system is developed, a method based on autocorrelation (AC) in conjunction with the discrete cosine transform (DCT) proposed for feature extractions from the pre-processed ECG signal. Also studied is the scenario where the proposed system deals with intruder signals in our database. For this goal, a study is performed to adjust the parameters allowing the system to avoid detection failure of false identification and false rejection scenarios. In addition, human brain activities represented by EEG are studied for biometric system purposes. In this study, an EEG-based biometric system is represented by performing a pre- processing stage on the EEG signals, with the features extractions completed using a wavelet packet decomposition and a classification.Item Study and simulation of the optical impairments in a CO-OFDM/RoF system(2016) Anbar, Lahcene; SadaliI, Soraya; ZitouniI, Abdelkader (Supervisor)One of the most advanced and attracting technologies in optical fiber communication systems that support efficient convergence of wireless and optical access network structure is Radio over Fiber (RoF). In RoF, light is modulated by using a radio signal and sent over an optical fiber link to simplify wireless access. As the demand for high-speed wireless communications is increasing rapidly, Coherent Optical Orthogonal Frequency Division Multiplexing (CO-OFDM) technology is considered to be a fundamental technique to satisfy this need. The aim of this project is to investigate through simulation the impairments due to the integration of CO-OFDM with RoF for achieving high data rates over long haul optical fiber systems. Solutions to combat the attenuation as well as the dispersion effects on such systems have been proposed and tested.