M. ZERDOUMI Zohra

MCA

Directory of teachers

Department

Departement of ELECTRONICS

Research Interests

Adaptive filtering linear and nonlinear equalization OFDM FBMC UFMC techniques 5G and 6G Digital communication system Artificial intelligence

Contact Info

University of M'Sila, Algeria

Email : zohra.zerdoumi@univ-msila.dz

On the Web:

  • Google Scholar N/A
  • ResearchGate
    ResearchGate N/A
  • ORCID N/A
  • SC
    Scopus N/A

Recent Publications

2023-11-04

The Morphological Analysis of Human Skin Layers Using Computational Image Segmentation

The human skin, being the largest organ of the body, plays a pivotal role in maintaining homeostasis and protecting the body from external threats. Understanding the morphology and composition of skin layers is essential for both medical and cosmetic applications. This study focuses on the morphological analysis of human skin layers utilizing advanced computational image segmentation techniques. The primary objectives of this research are twofold: first, to develop a robust computational image segmentation framework for accurately delineating the distinct layers of human skin, and second, to perform a comprehensive morphological analysis of these skin layers. The study leverages cutting-edge machine learning algorithms and image processing methodologies to achieve these objectives. To achieve accurate image segmentation, a combination of deep learning models and traditional image processing techniques is employed. Convolutional Neural Networks (CNNs) are trained on a dataset of high-resolution skin images to segment the epidermis, dermis, and subcutaneous layers. Post-processing steps, such as morphological operations and edge detection, are applied to refine the segmentation results.
Citation

M. ZERDOUMI Zohra, (2023-11-04), "The Morphological Analysis of Human Skin Layers Using Computational Image Segmentation", [international] 2023 IEEE International Workshop on Mechatronics Systems Supervision , Hammamet-Tunisia

Read Full Paper
2023-11-02

An improved learning algorithm for training neural network based lattice equalizer

In nonlinear channel equalization, artificial
neural networks (ANN) have attracted a significant interest. The
ANN's primary drawback is their intensive training. We
propose suggestions for enhancing their training capacities. The
first involves applying a whitening technique to the input data
by employing a lattice structure as the equalizer. Lattice
equalizer therefore becomes insensitivity to the inputs
correlation matrix. In the second strategy, we suggest modifying
the slope of the activation function. Combining the two methods
increases the ANN's nonlinear capabilities and adaptability.
Through simulation tests, the offered methodologies efficacy is
verified. The results demonstrate that the performance of the
neural network-based lattice equalizer is greatly improved by
whitening the received data using adaptive lattice channel
equalization techniques in conjunction with an adjustable slope
activation function.
Citation

M. ZERDOUMI Zohra, (2023-11-02), "An improved learning algorithm for training neural network based lattice equalizer", [international] 2023 IEEE International Workshop on Mechatronics Systems Supervision , Hammamet-Tunisia

Read Full Paper
2023-05-27

Parametric study of a triangular microstrip antenna with PBG substrate

The main objective of our work is to study the resonance characteristics of a triangular microstrip antenna with 2D Photonic Gap Band (PGB) substrate. We study the influence of different parameters of this antenna: the dimensions of the patch, PBG substrate height and permittivity, the diameter and the different networks of holes
on the resonance frequency, bandwidth and directivity, by using electromagnetic simulation tool in the frequency domain; CST based on the finite integration method
Citation

M. ZERDOUMI Zohra, BRIK FATIMA, DIB SAMIRA, , (2023-05-27), "Parametric study of a triangular microstrip antenna with PBG substrate", [national] First National Conference On Industrial Engineering And Sustainable Development CIESD’23 , University of Relizane, Algeria

Read Full Paper
2023-05-17

An Improved Recursive Least Square Algorithm For Adapting Fuzzy Channel Equalizer

Adaptive filters have been thoroughly investigated in digital communication. They are especially exploited
as equalizers, to compensate for channel distortions, although equalizers based on linear filters perform
poorly in nonlinear distortion. In this paper, a nonlinear equalizer based on a fuzzy filter is proposed and a
new algorithm for the adaptation parameters is presented. The followed approach is based on a
regularization of the Recursive Least Square (RLS) algorithm and an incorporation of fuzzy rules in the
adaptation process. The proposed approach, named Improved Fuzzy Recursive Least Square (IFRLS),
enhances significantly the fuzzy equalizer performance through the acquisition of more convergence
properties and lower steady-state Mean Square Error (MSE). The efficiency of the IFRLS algorithm is
confirmed through extensive simulations in a nonlinear environment, besides the conventional RLS, in
terms of convergence abilities, through MSE, and the equalized signal behavior. The IFRLS algorithm
recovers the transmitted signal efficiently and leads to lower steady-state MSE. An improvement in
convergence abilities is noticed, besides the RLS.
Citation

M. ZERDOUMI Zohra, (2023-05-17), "An Improved Recursive Least Square Algorithm For Adapting Fuzzy Channel Equalizer", [national] Engineering, Technology & Applied Science Research , ETASR

Read Full Paper
2022

Évitez le haut degré de similarité et écrire des citations de manière appropriée en utilisant Mendeley

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Citation

M. ZERDOUMI Zohra, (2022), "Évitez le haut degré de similarité et écrire des citations de manière appropriée en utilisant Mendeley", [international] Évitez le haut degré de similarité et écrire des citations de manière appropriée en utilisant Mendeley , Algérie,

Read Full Paper
2021-11-04

An Adaptive Sigmoidal Activation Function for Training Feed Forward Neural Network Equalizer

Feed for word neural networks (FFNN) have attracted a great attention, in digital communication
area. Especially they are investigated as nonlinear equalizers at the receiver, to mitigate channel distortions and
additive noise. The major drawback of the FFNN is their extensive training. We present a new approach to
enhance their training efficiency by adapting the activation function. Adapting procedure for activation function
extensively increases the flexibility and the nonlinear approximation capability of FFNN. Consequently, the
learning process presents better performances, offers more flexibility and enhances nonlinear capability of NN
structure thus the final state kept away from undesired saturation regions. The effectiveness of the proposed
method is demonstrated through different challenging channel models, it performs quite well for nonlinear
channels which are severe and hard to equalize. The performance is measured throughout, convergence
properties, minimum bit error achieved. The proposed algorithm was found to converge rapidly, and accomplish
the minimum steady state value. All simulation shows that the proposed method improves significantly the
training efficiency of FFNN based equalizer compared to the standard training one.
Citation

M. ZERDOUMI Zohra, Latifa ABDOU, Djamel BENATIA, , (2021-11-04), "An Adaptive Sigmoidal Activation Function for Training Feed Forward Neural Network Equalizer", [international] IConTech 2021: International Conference on Technology 4th to 7th November 2021, Antalya, Lara, Turkey , Antalya, Lara, Turkey

Read Full Paper
2021

cours Master2: Instrumentations et mesures industrielles

L'instrumentation est un domaine très vaste qui fait appel à de nombreuses
technologies, telles que : l’informatique, la vision artificielle, la régulation automatique.
L'instrumentation constitue une activité capitale en automatisme étant donné qu'elle fournit
les informations indispensables au contrôle des installations automatisées. L'exploitation des
informations délivrées par des mesures en instrumentation peuvent être facilitées par
des logiciels d'analyse du signal, de traitement et de visualisation de données. Dans ce
contexte, ce support de cours a pour objet de présenter un large éventail des connaissances de
base en instrumentation et mesure industrielle. Ce document est destiné aux étudiants de la
formation de master II instrumentation dans le cadre du programme officiel de l’enseignement
supérieur de la matière mesure et instrumentation industrielle.
Ce cours introduit des notions de base en mesure telles que les caractéristiques métrologique
(précision, résolution, temps de réponse, étendue de mesure, ...), les générateurs de tension (0-
10V), les générateurs d'intensité (0-20 mA et 4-20 mA) ainsi que les principes des appareils de
mesures analogique et numérique.
Ce document décrit aussi les maillons essentiels rencontrés dans une chaîne de mesure
analogique et numérique, justifiant ainsi leurs rôles .
Afin de donner un aperçu sur les systèmes de mesure industriels, ce support de cours présente
une variante d’exemples de procédés de mesure tels que : le tachymètre, le PH-mètre, le
débitmètre, les techniques spectrométriques ...
Etant donné qu’en instrumentation, le bruit affecte tous les éléments de la chaine de mesure, il
conditionne la précision de la mesure, il perturbe le fonctionnement des horloges, il génère des
fluctuations de fréquence celles-ci peuvent limiter le débit de transmission des données.
Citation

M. ZERDOUMI Zohra, (2021), "cours Master2: Instrumentations et mesures industrielles", [national] Mohammed Boudiaf Msila

Read Full Paper

An improved learning algorithm for nonlinear channel equalization

Abstract –Nonlinear channel equalization have attracted a great attention, in digital communication area. Especially artificial neural networks (ANN) are investigated as equalizers, to mitigate channel impairments. The major drawback of the ANN is their extensive training. We present a new learning approach to enhance their training efficiency by performing an adaptive activation function. The new learning procedure increases significantly the flexibility and the nonlinear approximation capability of ANN equalizer. The effectiveness of the proposed method is established via diverse challenging channel models. The proposed method is also performed on variant environment to check its tracking ability. Simulation results show that the proposed approach improves significantly the training efficiency of ANN based equalize compared to the standard training one.
Citation

M. ZERDOUMI Zohra, (2021), "An improved learning algorithm for nonlinear channel equalization", [international] 1st International Conference on Applied Engineering and Natural Sciences , Konya, Turkey

Read Full Paper
2018

Back propagation algorithm with adaptive slope sigmoidal activation function for mitigating channel impairments

We present a new approach to enhance their training efficiency by adapting the slope of the sigmoidal activation function. Adapting procedure for activation function extensively increases the flexibility and the nonlinear approximation capability of ANN. Consequently, the learning process presents a better performance; the final state kept away from undesired saturation regions. The effectiveness of the proposed method is demonstrated through different challenging channel models, in terms of convergence properties, minimum bit error achieved. The tracking capability of the proposed method is also considered.
Citation

M. ZERDOUMI Zohra, (2018), "Back propagation algorithm with adaptive slope sigmoidal activation function for mitigating channel impairments", [international] Second International Conference on Electrical Engineering (ICEEB’2018) , Biskra, Algeria

Read Full Paper
2016

An improved back propagation algorithm for training neural network-based equaliser for signal restoration in digital communication channels

Our approach consists on modifying the conventional back
propagation algorithm, through creating an adaptive nonlinearity in the activation function.
Experiment results evaluates the performance of the MLPE trained using the conventional BP
and the improved back propagation with adaptive gain (IBPAG). Due to the adaptability
of the activation function gain the nonlinear capacity and flexibility of the MLP is enhanced
significantly.
Citation

M. ZERDOUMI Zohra, (2016), "An improved back propagation algorithm for training neural network-based equaliser for signal restoration in digital communication channels", [national] Int. J. Mobile Network Design and Innovation, , inderscience

Read Full Paper

Multilayer Perceptron Based Equalizer with an Improved Back Propagation Algorithm for Nonlinear Channel

Neural network based equalizers can easily compensate channel impairments; such additive noise and
inter symbol interference (ISI). The authors present a new approach to improve the training efficiency
of the multilayer perceptron (MLP) based equalizer. Their improvement consists on modifying the back
propagation (BP) algorithm, by adapting the activation function in addition to the other parameters
of the MLP structure. The authors report on experiment results evaluating the performance of the
proposed approach namely the back propagation with adaptive activation function (BPAAF) next to
the BP algorithm.
Citation

M. ZERDOUMI Zohra, (2016), "Multilayer Perceptron Based Equalizer with an Improved Back Propagation Algorithm for Nonlinear Channel", [national] Int.J. of Mobile Computing and Multimedia Communications , IGI publisher

Read Full Paper
2015

‘Neural Networks Based Equalizer for Signal Restoration in Digital Communication Channels

This paper presents the equalization of
digital communication channels using artificial neural network structures. The performances of a
nonlinear equalizer using multilayer perceptron (MLP) trained by the back propagation
algorithm is compared with a conventional linear traversal equalizer (LTE). Simulation results
show that the performances of the MLP based Equalizer surpass significantly the classical LTE in
term of the restored signal, the steady state mean square error (MSE) achievable and the
minimum bit error rate attainable. The consistency in performance is observed in
minimum phase and non-minimum phase channels as well.
Citation

M. ZERDOUMI Zohra, (2015), "‘Neural Networks Based Equalizer for Signal Restoration in Digital Communication Channels", [national] International Letters of Chemistry, Physics and Astronomy , scipress

Read Full Paper
2013

Adaptive Decision Feedback Equalizer Based Neural Network for Nonlinear Channels

This paper investigates the application of artificial neural network to the problem of nonlinear channel equalization. The difficulties caused by channel distortions such as inter symbol interference (ISI) and nonlinearity can overcome by nonlinear equalizers employing neural networks. It has been shown that multilayer perceptron based equalizer (MLPE) outperform significantly linear equalizers. We present a multilayer perceptron based equalizer with decision feedback (MLP DFE) trained with the back propagation algorithm. The capacity of the MLP DFE to deal with nonlinear channels is evaluated. It is shown from simulation results that performance of the MLP DFE surpass significantly the MLPE in term of eye pattern quality, steady state mean square error (MSE), and minimum Bit Error Rate (BER).
Citation

M. ZERDOUMI Zohra, (2013), "Adaptive Decision Feedback Equalizer Based Neural Network for Nonlinear Channels", [national] The third International Conference on Systems and Control (ICSC’13), IEEE conf , Hotel Hilton, Algeria.

Read Full Paper

Adaptive Equalization of Digital Communication Channels Using Neural Network ,

One of the main obstacles to reliable communications
is the inter symbol interference (ISI). An adaptive equalizer is
required at the receiver to mitigate the effects of non-ideal channel
characteristics. The conventional way to combat with ISI is to
include an equalizer in the receiver. This paper presents the
equalization of communication channels using artificial neural
network structures. The performances of a nonlinear equalizer
using multilayer perceptron (MLP) trained by the back
propagation algorithm is compared with a conventional linear
traversal equaliser (LTE). Simulation results show that the
performances of the MLP Equalizer surpass significantly the
classical LTE in term of the equalized signal and the steady
state mean square error (MSE) achievable.
Citation

M. ZERDOUMI Zohra, (2013), "Adaptive Equalization of Digital Communication Channels Using Neural Network ,", [national] International Conference On Signal, Image, Vision And Their Applications( SIVA’13) , Guelma, Algeria.

Read Full Paper
← Back to Researchers List