Institut Supérieur d'Informatique de Mahdia

We study the vortices of energy minimizers in the London limit for the Ginzburg–Landau model with periodic boundary conditions. For applied fields well below the second critical field we are able to describe the location and number of vortices. Many of the results presented appeared in [H. Aydi, Doctoral Dissertation, Université Paris-XII, 2004], others are new. Résumé Nous étudions les tourbillons des minimiseurs de l'énergie de Ginzburg–Landau en supraconductivité dans la limite de London, et pour des conditions aux limites périodiques. Lorsque le champ magnétique appliqué est petit devant le second champ critique, nous décrivons le nombre et la localisation de ces tourbillons. Certains de ces résultats étaient présents dans [H. Aydi, Doctoral Dissertation, Université Paris-XII, 2004], d'autres sont nouveaux.

The aim of this work is to study the symmetric orthogonal polynomials’ sequences which are Dunkl-classical. We recover again that the generalized Hermite polynomials and the generalized Gegenbauer polynomials are the only symmetric Dunkl-classical through a different manner that in Ben Cheikh and Gaied [Characterization of the Dunkl-classical symmetric orthogonal polynomials. Appl Math Comput. 2007;187:105–114] by solving a differential-difference equation in the dual space of polynomials. Moreover, we deduce the structure relation that such a symmetric monic orthogonal polynomials satisfies.

In public transportation, simulation provides capabilities of investigation of complex interactions between the components of the transportation system, including infrastructure, vehicles, passengers and intelligent transportation system. Although many simulation platforms (either commercial or open source) exist to test, validate and evaluate the performance of control systems, the development of platforms that integrate specific requirements of public transport systems still requires much investigation and effort. This paper reviews the main features of traditionally used simulation platforms. A comparative analysis is provided based on different criteria related to infrastructure, vehicles or the ability to implement and test intelligent and distributed control architectures. Several research directions are also pointed out and discussed.

The rapid proliferation of the Internet of Things (IoT) has led to the interconnection of billions of intelligent sensors. However, this interconnection has also introduced significant security challenges. Recently, deep learning has shown promising results in several fields including attacks detection. This paper aims to improve IoT security through the application of deep learning techniques. Specifically, we chose the Convolutional Neural Network as a means to identify and counteract the most severe IoT attacks, such as denial-of-service (DOS) and distributed denial-of-service (DDoS) attacks. The experimental results demonstrate that our CNN is highly effective in identifying DDOS and DOS attacks in the real dataset Bot-IoT, achieving an accuracy rate of 99.920%.

The objective of this study is to analyze a model of competition for one resource in the chemostat with general interspecific density-dependent growth rates, taking into account the predator–prey relationship. This relationship is characterized by the fact that the prey species promotes the growth of the predator species which in turn inhibits the growth of the first species. The model is a three-dimensional system of ordinary differential equations. With the same dilution rates, the model can be reduced to a planar system where the two models have the same local and even global behavior. The existence and stability conditions of all steady states of the reduced model in the plane are determined according to the operating parameters. Using the nullcline method, we present a geometric characterization of the existence and stability of all equilibria showing the multiplicity of coexistence steady states. The bifurcation diagrams illustrate that the steady states can appear or disappear only through saddle-node or transcritical bifurcations. Moreover, the operating diagrams describe the asymptotic behavior of this system by varying the control parameters and show the effect of the inhibition of predation on the emergence of the bistability region and the reduction until the disappearance of the coexistence region by increasing this inhibition parameter.

Software-defined networking is an evolving network architecture beheading the traditional network architecture focusing its disadvantages in a limited perspective. A couple of decades before, programming and networking were viewed as different domains which today with the lights of SDN bridging themselves together. This is to overcome the existing challenges faced by the networking domain and an attempt to propose cost-efficient effective and feasible solutions. Changes to the existing network architecture are inevitable considering the volume of connected devices and the data being held together. SDN introduces a decoupled architecture and brings customization within the network making it easy to configure, manage, and troubleshoot. This paper focuses on the evolving network architecture, the software-defined networking. Unlike a generic view on the evolving network, which makes work as a review, this work addresses various perspectives of the architecture leaving it an intermediate work in between the review of the literature and implementation, contributing towards factors like the design, programmability, security, security behaviors, and security lapses. This paper also analyses various weak points of the architecture and evolves the attack vectors in each plane leaving a conclusion to further progress towards identifying the impacts of the attacks and proposing mitigation strategies.

In this paper, we propose a modified Hybrid Naïve Possibilistic Classifier (HNPC) for heart disease detection from the heterogeneous data (numerical and categorical) of the Cleveland dataset. The proposed classifier is based on a different pattern with regard to our former HNPC which have been recently proposed to deal with the same problem. As HNPC, the modified classifier separates data into two subsets (numerical and categorical) and then estimates possibility beliefs using the two versions of the probability-possibility transformation method of Dubois ets al. for numerical and categorical data, respectively. However, unlike HNPC which is based on two fusion steps to make decision from possibility estimations, our new classifier performs a common fusion to combine these beliefs. During this fusion, the product and the minimum as main combination operators for possibility measures are investigated. Experimental evaluations on the Cleveland dataset show that the proposed modified HNPC may outperform the former HNPC as well as the main classification techniques which have been used in recent related work.

This paper investigates a Hybrid Naïve Possibilistic Classifier (HNPC) to detect the presence of heart disease from the heterogeneous data (numerical and categorical) of the Cleveland dataset. The proposed classifier stands for the hybridization of two versions of Naïve Possibilistic Classifier (NPC) which have been recently applied on numerical and categorical data, respectively. To estimate possibility beliefs from data, each one of these two versions calls the probability-possibility transformation method of Dubois et al. Later, two fusion steps are performed to make decision. In the first fusion, possibility values are combined within each classifier using the product and the minimum operators for numerical and categorical data, respectively. Then, these two rules are investigated in the second fusion step to combine possibilities assigned to each class. The obtained results show that the proposed HNPC outperforms the main classification techniques which have been used in recent related work.

The traffic congestion has become a serious problem in a city. The traffic congestion had important consequences in terms of social, economic and environmental preoccupations. For this reason, several ITS was proposed and their role is to manage the existing highway, public transportation and railroad infrastructure to ease congestion and respond to crises. Developer of such system seek to have a system that insure a safer and more convenient travel for people. In this paper, we propose a system for junctions traffic lights control based on case based reasoning (CBR) approach and fuzzy sets theory. In fact, the CBR is always considered as a cyclic paradigm of Artificial Intelligence and that is used to learning and problem solving based on past experience. The developed system is tested with on a virtual junction and the obtained results are discussed.

<p style='text-indent:20px;'>We study an interspecific, density-dependent model of two species competing for a single nutrient in a chemostat, allowing for a predator-prey relationship between them. We have previously examined the system in the absence of species mortality, showing that multiple positive steady states can appear and disappear through a saddle-node or transcritical bifurcation. In this paper we include mortality. We give a complete analysis for the existence and local stability of all steady states of the three-dimensional system that cannot be reduced to two dimensional ones. Specializing the forms of the rate functions, we show how mortality destabilizes the positive steady state and that stable limit cycles emerge through supercritical Hopf bifurcations. To describe how the process behaves with respect to the choice of dilution rate and input concentration as control parameters, we determine the operating diagram theoretically and also numerically by using the software package MATCONT. The bifurcation diagram based on the input concentration shows various types of bifurcations of steady states, and coexistence either at a positive steady state or via sustained oscillations.</p>

<p style='text-indent:20px;'>A mechanistic model describing the anaerobic mineralization of chlorophenol in a three-step food-web is investigated. The model is a six-dimensional system of ordinary differential equations. In our study, the phenol and the hydrogen inflowing concentrations are taken into account as well as the maintenance terms. The case of a large class of growth kinetics is considered, instead of specific kinetics. We show that the system can have up to eight types of steady states and we analytically determine the necessary and sufficient conditions for their existence according to the operating parameters. In the particular case without maintenance, the local stability conditions of all steady states are determined. The bifurcation diagram shows the behavior of the process by varying the concentration of influent chlorophenol as the bifurcating parameter. It shows that the system exhibits a bi-stability where the positive steady state can lose stability undergoing a supercritical Hopf bifurcation with the emergence of a stable limit cycle.

The Industrial Internet of Things (IIo <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$T$</tex> ) is rapidly growing in tandem with security concerns. In this paper, we propose two deep learning models for classifying IIo <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$T$</tex> traffic in binary and multi-class contexts in order to detect intrusions in IIoT networks. To train the models, a recent public dataset is used. The results are very encouraging, with accuracy more than 99%.

Image registration is a crucial task in medical applications and is perceived as an optimization problem which has an important interest in clinical diagnosis. In this work, we propose an optimization strategy based on a specific design of genetic algorithm combined with the gradient descent optimizer within multi-resolution scheme. The performance of the proposed method was tested and evaluated on real multimodal registration scenarios from the Retrospective Image Registration Evaluation (RIR) database. Our method results were compared with those of existing registration methods, they are accurate and effective.

The main focus of this research is on the development of bimodal traffic controller for signposted road-rail junctions. The role of such systems is to manage the existing infrastructure to ease congestion and respond to crises. The system seek to have a system that insure a safer and more convenient travel for people. In this paper, the proposed controller is based on case based reasoning (CBR) approach. In fact, the CBR is always considered as a cyclic paradigm of Artificial Intelligence and that is used to learning and problem solving based on past experience. The developed system is tested with on a virtual junction and the obtained results are discussed.

Atlas-based segmentation is a high-level technique which provides highly accurate results particularly in the anatomical segmentation of medical images. The main idea of this technique consists in using a dataset of atlases which is perceived as a priori information necessary for providing the automatic segmentation of images. In the current state of the art of atlas-based techniques, the segmentation task is carried out based on a fixed database of atlases. In this article, we propose a multi-atlas segmentation method for brain MRI images based on a dynamic atlas database which is progressively updated each time a new case is added. The experiments show that the suggested technique seems to perform well compared to conventional multi-atlas based segmentation methods.

The objective of this study is to analyze a model of the chemostat involving the attachment and detachment dynamics of planktonic and aggregated biomass in the presence of a single resource. Considering the mortality of species, we give a complete analysis for the existence and local stability of all steady states for general monotonic growth rates. The model exhibits a rich set of behaviors with a multiplicity of coexistence steady states, bi-stability, and occurrence of stable limit cycles. Moreover, we determine the operating diagram which depicts the asymptotic behavior of the system with respect to control parameters. It shows the emergence of a bi-stability region through a saddle-node bifurcation and the occurrence of coexistence region through a transcritical bifurcation. Finally, we illustrate the importance of the mortality on the destabilization of the microbial ecosystem by promoting the washout of species. L'objectif de cette étude est d'analyser un modèle du chémostat impliquant la dynamique d'attachement et de détachement de la biomasse planctonique et agrégée en présence d'une seule ressource. En considérant la mortalité des espèces, nous donnons une analyse complète de l'existence et de la stabilité locale de tous les équilibres pour des taux de croissance monotones. Le modèle pré-sente un ensemble riche de comportements avec multiplicité d'équilibres de coexistence, bi-stabilité et apparition des cycles limites stables. De plus, nous déterminons le diagramme opératoire qui dé-crit le comportement asymptotique du système par rapport aux paramètres de contrôle. Il montre l'émergence d'une région de bi-stabilité via une bifurcation noeud col et l'occurrence d'une région de coexistence via une bifurcation transcritique. Enfin, nous illustrons l'importance de la mortalité sur la déstabilisation de l'écosystème microbien en favorisant le lessivage des espèces.

This paper addresses a worker assignment problem where multi-skilled workers perform a set of tasks independently and without walking and work sharing between workstations. A real case study is considered in this work. This problem is solved as a non-identical parallel machines scheduling problem with stochastic processing time where each machine represent one worker. A simulation optimization (SO) approach, using a robust tabu search (TS) algorithm that is coupled with a simulation model, is proposed to minimize the maximum completion time (makespan). The results is compared with this obtained by the using assignment method in the SETA Company.

The advent of 5G and beyond networks is envisioned to support lower latency, higher data rates, and wider connectivity than previous cellular network generations. However, given the denser deployment of base stations (BSs) to accommodate such improvements, this results inevitably in a significant and unsustainable increase in the network’s energy consumption. Sleep Control (SC), which allows switching off some BS hardware components during light-traffic time, is considered a viable solution for greener and more energy-efficient Radio Access Networks (RAN). However, the optimization of SC is a highly challenging large-scale network combinatorial problem that depends on dynamic wireless channel conditions and varying traffic demands with stringent Quality-of-Service (QoS) requirements. Driven by the benefits and efficiency of Deep Reinforcement Learning (DRL), which has been successfully applied to multiple wireless network optimization problems, this paper investigates DRL approaches addressing sleep control in 5G and beyond RAN. To this end, we propose a taxonomy to classify the related literature. Then, we provide an overview of the different components of the Markov Decision Process (MDP) modeling the sequential decision-making of sleep control and the applied DRL algorithms. Finally, we highlight the main challenges in existing works and suggest novel strategies to address them.

To guarantee convergent state estimates and exact approximations, it is highly desirable that observers can independently dominate the effects of unmodelled dynamics. Based on adaptive nonlinear approximation, this paper presents a robust H ∞ gain neuro-adaptive observer (R H ∞ GNAO) design methodology for a large class of uncertain nonlinear systems in the presence of time-varying unknown parameters with bounded external disturbances on the state vector and on the output of the original system. The proposed R H ∞ GNAO incorporates radial basis function neural networks (RBFNNs) to approximate the unknown nonlinearities in the uncertain system. The weight dynamics of every RBFNN are adjusted online by using an adaptive projection algorithm. The asymptotic convergence of the state and parameter estimation errors is achieved by using Lyapunov cogitation under a well-defined persistent excitation condition, and without recourse to the strictly positive real condition. The repercussions of unknown disturbances are reduced by integrating an H ∞ gain performance criterion into the proposed estimation approach. The condition imposed by this proposed observer approach, such that all estimated signals are uniformly ultimately bounded, is expressed in the form of the linear matrix inequality problem and warrants the demanded performances. To evaluate the performance of the proposed observer, various simulations are presented.

International audience