African Review of Science, Technology and Development

Table of Content

Fri, 31 Dec 2021 00:00:00 +0100

Table of Content

Support Vector Machines and Convolutional Descriptors for Arabic Handwritten Character Recognition

Mamoun Mamouni, Djamaaoui Mebrouka, Zembou Meriem — Fri, 31 Dec 2021 00:00:00 +0100

The field of handwriting recognition is still a great challenge and competition between researchers; in this work, we construct a system for recognition of Arabic handwritten characters using support vector machines (SVM). Our system includes the steps: preprocessing, feature extraction, and classification. In the feature extraction phase, we simulate the behavior of convolutional neural networks (CNN) by integrating convolutional descriptors, where we used five filters: Prewitt, Sobel, Laplacian, point and line detection filter followed by a data reduction step using the Zoning method. To evaluate our approach, we created our own database of Arabic handwritten characters containing 1745 images; also, we used the HACDB database to test our system, the results we have achieved are very encouraging.

Artificial Immune System Based Multi Hop Energy Efficient Routing Protocol For Wireless Sensor Network

Mohammed Demri, Khaled Bouafia — Fri, 31 Dec 2021 00:00:00 +0100

Clustering is one of the important methods for prolonging the network lifetime in wireless sensor networks (WSNs). It involves grouping of sensor nodes into clusters and electing cluster heads (CHs) for all the clusters. CHs collect the data from respective cluster’s nodes and forward the aggregated data to base station. A major challenge in WSNs is to select appropriate cluster heads. In this work, we present a new algorithm for Cluster-Head selection, which is based on artificial immune system (AIS) with a new two-hop routing technique for serving remote nodes using the vice-Cluster Head plus a multi-hop routing technique. To confirm the improvements brought by our algorithm we conducted a simulation using the NS2 network simulator, in which the performance of our algorithm are evaluated and compared with the existing clustering protocol LEACH.

Pure and Hybrid ARQ Mechanisms: Comparison and Analysis

Mbarka Bencheikh, Halima Targui, Mohammed Omari — Fri, 31 Dec 2021 00:00:00 +0100

In recent years, wireless multimedia applications have gained the attention of researchers due to the rapid growth in wireless technology and the Internet. TCP (Transmission Control Protocol) and its variants such as TFRC (TCP Friendly Rate Control) over wireless links have been widely studied to be suitable for wireless multimedia applications, as TCP is the most widely used transport protocol in the Internet. current (approximately 90 percent of the overall IP traffic is carried by TCP according to). When TCP connections span over wireless links, many factors such as interference, multipath fading, user mobility, and atmosphere can cause errors resulting in dropped frames. In this work, we are studying different techniques of error control mechanisms of type ARQ and obtaining an analysis of their performance.

Finding a Shorter Path in Wireless Sensor Networks

Mohammed Kaddi, Siham Ferouhat, Imane Sarhani — Fri, 31 Dec 2021 00:00:00 +0100

The objective of operational research is to find optimal solutions to economic problems using mathematical methods capable of being programmed by computer. Graph theory is a fundamental part of operations research. The objectives of our work is to study the problem of finding a shorter path by applying Dijkstra's algorithms in wireless sensor networks. In this work, we tried to shorten the path of transmitted data packets since their energy consumption is about 70% of the overall energy consumption of the network.

Integration of an Arabic Lemmatizer as Part of an Information Retrieval System

Mohamed Amine Cheragui, Zoulikha Benblal, Fatima Belouafi — Fri, 31 Dec 2021 00:00:00 +0100

Natural Language Processing (NLP) is a field of Artificial Intelligence that encompasses several topics: Machine translation, Text summarization, spell checkers, information retrieval, etc. The objective of information retrieval (IR) is to provide a user with easy access to the information of interest, this information being located in a mass of textual documents. To achieve this objective, it must represent, store and organize information, then provide the user with the elements corresponding to the information need expressed by his request. Most of information retrieval systems (IRS) use simple terms to index and retrieve documents based on models such as: Boolean model, vector model, probabilistic model, language model, etc. The aim of our work is to develop a hybrid Arabic system based on four (04) models where the originality of the work lies in incorporating a stemmer in the search process to improve the results of our system.