Ahmed Fakhfakh

Comparison of SPWM and SVPWM control of electrical vehicle in VHDL-AMS

This paper presents a VHDL-AMS description of a Space Vector Pulse Width Modulation (SVPWM) and Sinusoidal Pulse Width Modulation control of a permanent magnet synchronous motor (PMSM) and exposes some high-level simulation results allowing a behavioral study with a virtual way. We compared between these two methods to optimize the response time of the Electric Vehicle.

Modeling and simulation of electrical vehicle in VHDL-AMS

Regarding limitation of fossil fuels and the high consumption rate of this energy for transportation, inclination of vehicle industry toward other sources of energy is inevitable. Electric vehicles and hybrid vehicles could be a good solution. Thanks to the state of art of electric motors, power electronics, embedded power train controller and energy storage systems like batteries, the performance of the vehicle could become more and more energy efficient. Since the integrating of all these components in a drive train configuration could be a challenge for the manufacturer, computer simulation and modeling before prototyping could be really beneficial in terms of cost, safety and design performance. In this paper some of the principles of modeling and simulation are discussed. As traction chain plays a significant role in electric vehicles, its modeling and optimization has a significant importance. As an electric vehicle is a multidisciplinary system, the new standard VHDL-AMS since 1999 is suitable for its modelling and simulation in the same software environment and with different abstraction levels.

Performance analysis of correlation filters used for face recognition

There are many approaches for designing filters to be used with optical correlation techniques that presents a good method to carry out instantly a decision on the target form. In this paper, we are interested on filters designs since their important role for target tracking and identification. Considering a Vanderlugt correlator (VLC), we discuss the basis of existing filters in the literature, and we carry out a performance analysis of the above filters in terms of Receiver Operating Characteristic (ROC) and peak-to-correlation energy (PCE). Simulation results prove that POF is more appropriate for optical correlator to be used for face recognition.

Integrated wide-band FSK demodulator based on ISPD PLL

Classic designs of FSK demodulator generally employ conventional phase locked loop second order (PLL). In this paper, we explore the design of a new architecture (without using any filters) of high-performance integrated FSK demodulator using ISPD PLL (inverse sine phase detector PLL). This architecture has clearly shown superior frequency performances, such as large bandwidth, large keep range, large seizure range and a perfect reconstruction of an input signal, when compared to the conventional architecture. The digital design of the system involves the use of the SIMPLORER simulator for the hardware description language VHDL-AMS.

A new correlation criterion based on energy distribution: face recognition application

Today, communications security, i.e. the discipline of preventing unauthorized interceptors from accessing telecommunications in an intelligible form, while still delivering content to the intended recipients, is a main issue in our modern society especially. In this paper, attention is drawn to the importance and relevance of optical correlation techniques for detection and tracking people. In order to be efficient, these techniques need pre- or post-processing steps to take into account the environmental conditions. The aim of this work is to improve the performance of the optical correlation method, based on a new decision process in order to reduce the false detection rate. To realize this, we propose a method using a VanderLugt correlator with a phase-only filter for face recognition using two criteria for decision making based on the values of the peak-to-correlation energy and the energy distribution in different parts of the correlation plane. In the three-step algorithm, the first stage consists by dividing the correlation plane into nine equal sub-planes. In the second stage the energy of each sub-plane is computed, while in the last stage the classification criterion is realized and the recognition rate is calculated. Numerous tests were performed using the Pointing Head Pose Image Database. They show the effectiveness of the method in terms of face recognition detection rate without pre-processing phase and with 0% false detection.Today, communications security, i.e. the discipline of preventing unauthorized interceptors from accessing telecommunications in an intelligible form, while still delivering content to the intended recipients, is a main issue in our modern society especially. In this paper, attention is drawn to the importance and relevance of optical correlation techniques for detection and tracking people. In order to be efficient, these techniques need preor post-processing steps to take into account the environmental conditions. The aim of this work is to improve the performance of the optical correlation method, based on a new decision process in order to reduce the false detection rate. To realize this, we propose a method using a VanderLugt correlator with a phase-only filter for face recognition using two criteria for decision making based on the values of the peak-to-correlation energy and the energy distribution in different parts of the correlation plane. In the three-step algorithm, the first stage consists by dividing the correlation plane into nine equal sub-planes. In the second stage the energy of each sub-plane is computed, while in the last stage the classification criterion is realized and the recognition rate is calculated. Numerous tests were performed using the Pointing Head Pose Image Database. They show the effectiveness of the method in terms of face recognition detection rate without pre-processing phase and with 0% false detection. Pattern Recognition and Tracking XXIX, edited by Mohammad S. Alam, Proc. of SPIE Vol. 10649, 1064911 ·

Joint pre-processing phase to optical correlator for face recognition applications

In this work, our main objective is to enhance the performance of the optical correlator which is VunderLaugt (VLC) correlator method for face recognition applications. We propose a pre-processing phase to improve robustness of VLC correlator for face recognition applications. Our technical results prove that this method is remarkably efficient to enhance the performances of a VLC correlator. We have used many tests based on Pointing Head Pose Image (PHPID), and we have considered faces through horizontal and vertical rotations. We have demonstrated that the pre-processing phase ameliorate the performance of the VLC correlator technique for face recognition applications. These performances are conducted from The receiving operating characteristics (ROC) curves.

Modeling from an Object and Multi-object Tracking System

This paper describes an implementation system tracking and detection object and multi-object. We trait the different tracking algorithms of moving objects in video sequences. The tracking a particular precise object, robust, efficient, reliable and fast is a difficult problem to solve. Hence, it is a very essential for computer vision applications. For this reason, tracking objects in video sequences is a very active area of research since the 1970. It has attracted many people in the area of research for use in a variety of applications. The tracking is often the first step in an analysis of the activities, detection, behavior, interactions and relationships between objects of interest. Many methods of tracking objects have been proposed and developed. Tracking is the estimation and analysis of the trajectories of an object in the frame plane by moving in an frame where video sequence. Most of the motion objects tracking algorithms take input frames captured by a fixed camera to give at output a video sequence. However, there are other algorithms that take as input a video sequence to provide a video at output containing the tracked objects. These algorithms perform a first step of the detection objects in order to determine which of the pixels of the current frame which belong to the background of the sequence and which represent the motion objects. The problem of object tracking can be expressed in terms of detection of the object in each frame of the video sequence. The set of applications (security systems, military systems, intelligent transportation systems, video conferencing, surveillance, etc.) can be addressed. So, the lightness of the systems also lets you open recent applications such as human-machine interface and/or the human-robot interface, where the sensors can be embedded on a mobile robot. This usually requires the exploitation of object tracking techniques and algorithms in motion that will register and implement on FPGA targets that are introduced in these interfaces. We can cite as an example then the main existing algorithms in the literature as the block-matching, the KLT, the Kalman filter, the Meanshift and the Camshift.

Implementation of optical correlator for face recognition applications

The main purpose of this work is to put together the benefits of the optical correlator which is VunderLaugt (VLC) correlator method for face recognition applications. With this aim in mind, we compare the performances of VLC correlator based on the fast Fourier transform (FFT) with the optical Fourier Transform (NO_FT) based on Fraunhofer diffraction. To achieve this goal, numerical implementation of the optical FT is needed. In this paper, we suggest and validate an all-numerical implementation of a VLC correlator with optical FT. Different tests using the Pointing Head Pose Image Database (PHPID) and considering faces with vertical and horizontal rotations were performed. The receiving operating characteristics (ROC) curves show that the optical FT simulating the Fraunhofer diffraction leads to better performances than the FFT.

Fast and accurate behavioural simulation of fractional-N frequency synthesizer for the optimization of the lock time

Today, the current need consisting of implementing more and more complex systems imply the implementation of new methodologies to make the CAD product reliable in order to improve time to market, study costs, reusability and reliability of the design process. This paper proposes a high level design approach applied for the simulation and the optimization of fractional-N synthesizer acting as a direct GPSK modulator and designed for the UMTS standard application. It uses the hardware description language VHDL-AMS and a genetic algorithm to optimize the modulator with a considerably reduced CPU time.

High level characterization and optimization of a GPSK modulator with genetic algorithm

Today, design requirements are extending more and more from electronic (analogue and digital) to multidiscipline ones. These current needs imply implementation of methodologies to make the CAD product reliable in order to improve time to market, study costs, reusability and reliability of the design process. This paper proposes a high level design approach applied for the characterization and the optimization of fractional-N synthesizer acting as a direct GPSK modulator and designed for the UMTS standard application. It uses the hardware description language VHDL-AMS and a genetic algorithm to optimize the modulator with a considerably reduced CPU time before passing to a transistor level characterization.