Oujda, Morocco
Oujda, Morocco

Time filter

Source Type

Benhamed C.,University of Science and Technology Houari Boumediene | Mekaoui S.,University of Science and Technology Houari Boumediene | Ghoumid K.,Complexe University
International Journal of Design and Nature and Ecodynamics | Year: 2016

An SVM (Support Machine Vector) algorithm has been implemented to sense traffic anomalies through a large-scale IP Network. We have applied this algorithm on data provided by the well-known large-scale American IP Network (Abilene Network). The developed SVM algorithm can classify the Network traffic into two categories of classes namely: normal; and abnormal. The implementation of this algorithm has been performed on real collected data thanks to Netflow protocol and has yielded satisfactory results with a classification rate going over 96% and a false alarms rate lower than 10%. © 2016 WIT Press.

Mekaoui S.,University of Science and Technology Houari Boumediene | Tchoketchkebir S.,University of Science and Technology Houari Boumediene | Ghoumid K.,Complexe University
WIT Transactions on Biomedicine and Health | Year: 2013

The aim of the work presented in this paper is focused on the super-resolution technique for image processing in order to measure efficiently the diameter of the human artery. This work can find crucial applications in avoiding cerebral aneurisms if the physician has a good monitoring tool that can allow him to get this information early on the basis of the analysis of cerebral artery images. For this purpose, we have used a simulated artery home probe made of silicon and this model had been scanned by Phillips flat panel scanner (ALURA FD 20) with a resolution of 0.035mm. We have developed software based on the superresolution algorithm using the multi-scale wavelet analysis and able to reconstruct a high resolution image closest to the reality from a low resolution image. We have applied our image processing software to many images and have carried out a comparison with a super-resolution technique based on polynomial interpolation or B-splines interpolation and find out that our method yields better measurements of the artery diameter. © 2013 WIT Press.

Benmakhlouf T.,University of Science and Technology Houari Boumediene | Mekaoui S.,University of Science and Technology Houari Boumediene | Ghoumid K.,Complexe University
WIT Transactions on Information and Communication Technologies | Year: 2013

The main aim of this paper is concerned with modern telecommunications systems which involve modern methods of coding, encryption and decryption of speech signals. For a long time, for the transmission of a speech signal analog telecommunications systems have been used. Because of unexpected and unavoidable interference, wave fading perturbations and different kinds of noise occurring in the channel, it was not possible to detect and receive the same transmitted speech signal. Consequently, digital systems have steadily replaced the former. Here, we have simulated two blocks of such systems, namely the source coding block and the encryption/decryption block. We tested them by listening to the synthesized signals via headphones and using a simulation operated using Simulink of the source software Matlab. Although metallic in their tonalities, results were found to be acceptable. © 2013 WIT Press.

Ghoumid K.,Complexe University | Ghoumid K.,University of Franche Comte | Elhechmi I.,University of Franche Comte | Mekaoui S.,University of Science and Technology Houari Boumediene | And 2 more authors.
Optics Communications | Year: 2013

In this paper the authors present an hybrid approach for the analysis of the optical filtering function in corrugated waveguide filters with a high index modulation. This approach is based on the hybridization of the extended couple mode theory (CMT) with the transfer matrix approach. The authors chose to treat the case of high index corrugation because in this case the theory elaborated before is not rigorously applicable. The proposed approach allows the calculation of the reflection coefficient and the full width at half maximum (FWHM) for any index modulation scheme. The hybridization of both methods mentioned above explains the impact and effects of opto-geometric parameters on the reflection coefficient and the bandwidth at mid-height. The theoretical results are verified by experimental measurements realized on Ti:LiNbO 3 waveguides with a high index modulation experimentally implemented by engraving using Focused Ion Beam (FIB) process. © 2012 Elsevier B.V.

Benkhaldoun F.,University of Paris 13 | Daoudi S.,Complexe University | Elmahi I.,Complexe University | Seaid M.,Durham University
Mathematics and Computers in Simulation | Year: 2011

Unstructured finite volume methods are receiving increased attention mainly because of their ability to provide a flexible spatial discretization. Hence, some areas can be resolved in great detail while not over-resolving other areas. Development of these models is an ongoing process with significant longstanding issues with adaptive grids, efficiency, well-balanced flux-gradient and source-term approximations, and so forth. However, many of these problems have been solved with the current generation of models which have much promise for morphodynamic models in contracting channel flows. Our purpose is to compare a class of unstructured finite volume morphodynamic models in contracting channel flows. The proposed adaptive finite volume method has important advantages in the discretization of the gradient fluxes and source terms and can handle complex topography using unstructured grids and satisfies the conservation property. The adaptation criteria is based on monitoring the bedload in the computational domain during its transport process. The comparison is illustrated with a simple test case of contracting channel flow. © 2011 IMACS.

Chaabelasri E.M.,University Mohammed Premier | Salhi N.,University Mohammed Premier | Elmahi I.,Complexe University | Benkhaldoun F.,LAGA
Physical and Chemical News | Year: 2010

This work concerns the numerical simulation of free flow on irregular bed. This flow can be described by the shallow water or Saint-Venant equations, written in conservative form. The numerical approximation model is based on non structured finite volume method coupled with dynamical adaptive mesh for spatial discretisation and a second order TVD Runge-Kutta scheme for temporal integration. The numerical flow at interface of each volume was evaluated by Roe approximate Riemann solver, extensible in second order by MUSCL approach. The source term was discretised by upwinding scheme of Vazquez. The resultant scheme is non-oscillatory and possesses conservation property required by the well balanced schemes. The precision, robustness and benefics in CPU time have been demonstrated over numerical tests including the principal's hydraulics problems such as: flood wave propagation, hydraulic jump and transcritical flow over variable bed.

Chaabelasru E.,University Mohammed Premier | Borthwick A.G.L.,University of Oxford | Salhi N.,University Mohammed Premier | Elmahi I.,Complexe University
World Journal of Modelling and Simulation | Year: 2014

A balanced adaptive scheme is proposed for the numerical solution of the coupled non-linear shallow water equations and depth-averaged advection-diffusion pollutant transport equation. The scheme uses the Roe approximate Riemann solver with centred discretization for advection terms and the Vazquez scheme for source terms. It is designed to handle non-uniform bed topography on triangular unstructured meshes, while satisfying the conservation property. Dynamic mesh adaptation criteria are based on the local pollutant concentration gradients. The model is validated for steady flow over irregular bed topography, recirculation due to a sidewall expansion in a frictionless channel, and pollution advection in a flat-bottomed channel. An idealised application to the simulation of pollution dispersion in the Bay of Tangier, Morocco is presented, which demonstrates the capability of the dynamically adaptive grid model to represent water quality scenarios in a bay of non-uniform bed topography and complicated shoreline.

Mekaoui S.,H+ Technology | Benhamed C.,H+ Technology | Ghoumid K.,Complexe University
WIT Transactions on Information and Communication Technologies | Year: 2013

This paper deals with a method using a specific class of neural networks whose learning phase is based on the Levenberg-Marquardt algorithm and which had been applied to the estimation of the traffic matrix (TM) of a large scale IP network. The neural network had been implemented with the help of the specific neural toolbox of the source software Matlab. Such neural networks are within the class of feed forward and recurrent types. The simulation tests have been processed on the available data base of the very reputed American observatory data base on the Internet of a very large scale IP network, the so-called Abilene network on both categories of neural networks. The simulated results using this method have been found to be very accurate as compared to one another. The static model converges rapidly but was less accurate in the estimation of the Traffic Matrix of such a kind of large IP System (the Abilene System) than the dynamic model which in this way earned the challenge of yielding a perfect estimation. © 2013 WIT Press.

Rahmoune M.,Complexe University | Osmont D.,ONERA
Mechanika | Year: 2010

Several authors have focused on the construction of the adapted or specific finite elements to take into account the electromechanical coupling of piezoelectric materials, in adding the electric degrees of freedom DOF. The computation times can be expensive in this case. Some authors have eliminated the electric potential DOF by "compensation by condensation". It can justify a gain computation time. This work shows that the classic finite elements (without electrical DOF) are sufficient for solving a thin piezoelectric smart structure. The classic finite elements (mechanical only) are sufficient, we must just introduce correctly the modified elasticity matrix and the electric force induced from imposed potential. The electric load is computed from the obtained displacement. We have shown that the modified terms are the same as those used in the "compensation by condensation" method. We have avoided the computation times by "compensation by condensation" steps and sensibly reduced the computation time.

Abdellaoui R.,Complexe University | Elmahi I.,Complexe University
Physical and Chemical News | Year: 2014

In this work, a Reynolds Averaged Navier-Stokes (RANS) model is adopted in a finite volume method for the numerical computation of turbulent shallow water flows on unstructured meshes. The governing equations are those of 2D Saint-Venant, they include the slope variations, the friction terms and the eddy viscosity. The effects of turbulence are incorporated in the system by considering the k-ε model. The convective fluxes are approximated using a modified Roe scheme that incorporates in its reconstruction the sign of the Jacobian matrix of the system. The diffusion terms are discretized using a Green-Gauss type reconstruction. The performance of the solver is illustrated over two test cases: the mixing layer problem in shallow water flows and the flow in an open channel with a bend of 90°. ©(2014), E-flow PDF Best Edition. All rights reserved.

Loading Complexe University collaborators
Loading Complexe University collaborators