Abdelouhab Zeroual

Identification and Prediction of Internet Traffic Using Artificial Neural Networks

This paper presents the development of an artificial neural network (ANN) model based on the multi-layer perceptron (MLP) for analyzing internet traffic data over IP networks. We applied the ANN to analyze a time series of measured data for network response evaluation. For this reason, we used the input and output data of an internet traffic over IP networks to identify the ANN model, and we studied the performance of some training algorithms used to estimate the weights of the neuron. The comparison between some training algorithms demonstrates the efficiency and the accu-racy of the Levenberg-Marquardt (LM) and the Resilient back propagation (Rp) algorithms in term of statistical crite-ria. Consequently, the obtained results show that the developed models, using the LM and the Rp algorithms, can successfully be used for analyzing internet traffic over IP networks, and can be applied as an excellent and fundamental tool for the management of the internet traffic at different times.

Blind parametric identification of non-Gaussian FIR systems using higher order cumulants

Two approaches are introduced for the identification of linear time-invariant systems when only output data are available. The input sequences are independent and must be non-Gaussian. To estimate the parameters of the system, we use only the fourth-order cumulants of the output, which may be contaminated by an additive, zero mean, Gaussian noise of unknown variance. To measure the performance of the proposed algorithms against existing methods, we compared them with the Zhangs algorithm. Simulations verify an apparent performance of the second algorithm, compared with the first and Zhangs algorithms, in a low signal-to-noise ratio and for small data. The simulation results show that the first algorithm has the same performance compared with Zhangs one. But the second algorithm achieves better performance compared with the first and Zhang is algorithm. For validation purposes, the second algorithm is used to search for a model able to describe and simulate the data set representing the wind speed.

ANFIS method for forecasting internet traffic time series

In This paper we have applied the adaptive neuro-fuzzy inference system (ANFIS) which is realized by an appropriate combination of fuzzy systems and neural networks for forecasting a set of input and output data of internet traffic time series. Several statistical criteria are applied to provide the effectiveness of this model. The obtained results demonstrate that the ANFIS model present a good precision in the prediction process of internet traffic in terms of statistical indicators. This model fits well real data and provides an effective description of network condition at different times.

Review article: Identification of quadratic systems using higher order cumulants and neural networks: Application to model the delay of video-packets transmission

This work concerns the development of two approaches for the identification of diagonal parameters of quadratic systems from only the output observation. The systems considered are excited by an unobservable independent identically distributed (i.i.d), stationary zero mean, non-Gaussian process and corrupted by an additive Gaussian noise. The proposed approaches exploit higher order cumulants (HOC) (fourth order cumulants) and are the extension of the algorithms developed in the linear version 1D, which uses a non-Gaussian signal input. For test and validity purpose, these approaches are compared to recursive least square (RLS), least mean square (LMS) and neural network identification algorithms using non-linear model in noisy environment. To demonstrate the applicability of the theoretical methods on real processes, we applied the developed approaches to search for models able to describe the delay of the video-packets transmission over IP networks from video server. The simulation results show the correctness and the efficiency of the developed approaches.

Identification of Quadratic Non Linear Systems Using Higher Order Statistics and Fuzzy Models

In this work we compare tow methods for the identification of non-linear systems. The first one uses a quadratic non linear model of which parameters are estimated using a new algorithm based on the fourth order cumulants. The second one is based on the Takagi-Sugeno fuzzy models. The simulation results show that the fuzzy models give the good results in noiseless and weak noise environment. However the quadratic model of which parameters are identified using the proposed algorithm works well in the high noise environment case

Gene-expression-based cancer classification through feature selection with KNN and SVM classifiers

This paper presents a study of feature selection methods effect, using a filter approach, on the accuracy and error of supervised classification of cancer. A comparative evaluation between different selection methods: Fisher, T-Statistics, SNR and ReliefF, is carried out, using the dataset of different cancers; leukemia cancer, prostate cancer and colon cancer. The classification results using k nearest neighbors (KNN) and support vector machine (SVM) classifiers show that the combination between SNRs method and the SVM classifier can present the highest accuracy.

Modeling non linear real processes with ANN techniques

In this paper we are interested to model real process by applying artificial neural networks technique. The performance and the aptitude of two types of this technique (multilayer perceptron neural network (MLP) and a radial basis function neural network (RBF)) are compared and applied for identifying non linear system of internet traffic network processes. In term of statistical criteria, the obtained results show the advantages of the developed model based on the RBF to describe the internet traffic.

Blind identification of non-linear quadratic systems using higher order cumulants and non-Gaussian input signals

This work concerns the problem of the identification of the kernels of non-linear quadratic systems using cumulants of the output data corrupted by a Gaussian noise, when the input is a stationary zero mean non-Gaussian white stochastic process. The proposed approach constitutes an extension of linear systems identification algorithm to non-linear quadratic systems using third-order cumulants. The developed algorithm is tested and compared with a recursive least square and a least mean square methods using different quadratic models for various values of signal to noise ratio and different sample sizes N. The simulation results show the efficiency and the accuracy of the proposed algorithm in non-linear quadratic system identification.

Cancer Classification Using Gene Expression Profiling: Application of the Filter Approach with the Clustering Algorithm

In this paper, we investigate the classification accuracy of different cancers based on microarray expression values. For this purpose, we have used hybridization between a filter selection method and a clustering method to select relevant features in each cancer dataset. Our work is carried out in two steps. First, we examine the effect of the filter selection methods on the classification accuracy before clustering. The studied filter selection methods are SNR, ReliefF, Correlation Coefficient and Mutual Information. The K Nearest Neighbor, Support Vector Machine and Linear Discriminant Analyses classifier were used for supervised classification task.

Design of a miniaturized Microstrip Patch Antenna for a passive UHF RFID tag

This paper present a design and miniaturization of rectangular patch antenna with balanced feed. The miniaturized patch antenna for a passive radio frequency identification (RFID) tag witch can operate in the ultra-high frequency (UHF), the resonant frequency is 915MHz. The simulation was performed in High Frequency Structure Simulator Software (HFSS). The miniaturized antenna has an acceptable results in terms of efficiency, size, return loss (S11), and Input Impedance.