TAHA HOCINE Kerbaa
طه حسين كربع
tahahocine.kerbaa@univ-msila.dz
0698193757
- Departement of ELECTRONICS
- Faculty of Technology
- Grade PHd
About Me
Science et Technologies
Research Domains
Radar Signal Processing
FiliereElectronique
Electronics andTelecommunications
Location
Msila, Msila
Msila, ALGERIA
Code RFIDE- 1989-10-28 00:00:00
-
TAHA HOCINE Kerbaa birthday
- 2023
-
2023
Multi-headed deep learning-based estimator for correlated-SIRV Pareto type II distributed clutter
This paper deals with the problem of estimating the parameters of heavy-tailed sea clutter in high-resolution radar, when the clutter is modeled by the correlated Pareto type II distribution. Existing estimators based on the maximum likelihood (ML) approach, integer-order moments (IOM) approach, fractional-order moments (FOM), and log-moments (log-MoM) have shown to be sensitive to changes in data correlation. In this work, we resort to a deep learning (DL) approach based on a multi-headed architecture to overcome this problem. Offline training of the artificial neural networks (ANN) is carried out by using several combinations of the clutter parameters, with different correlation degrees. To assess the performance of the proposed estimator, we resort to Monte Carlo simulation, and we observed that it has superior performance over existing approaches in terms of estimation mean square error (MSE) and robustness to changes of the clutter correlation coefficient.
Citation
Taha hocine Kerbaa , , (2023), Multi-headed deep learning-based estimator for correlated-SIRV Pareto type II distributed clutter, EURASIP J. Adv. Signal Process, Vol:81, Issue:81, pages:25, Springer Nature
- 2022
-
2022
On the Performance of GLRT-LTD CFAR Processor in Correlated Pareto Clutter Under Different Estimators
Pareto type II distribution is a class of high-resolution sea-reverberation data models. Application of the GLRT-LTD (Generalized Likelihood Ratio Test Linear Threshold Detector) algorithm requires an accurate estimation of the clutter parameters. Under the assumption of correlated Pareto clutter, several estimators could be applied. In this work, we investigate the effect of the MLE (Maximum likelihood Estimation), Integer order moments, fractional-order moments, and zlog(z) estimators on the detection performance of the GLRT-LTD procedure. From simulated datasets, it is shown that approximate results are obtained by MLE and zlog(z) methods. Moreover, the zlog(z) approach is advantageous when complicated parameter estimation scenarios occur (i.e., correlation coefficient tends to one).
Citation
Taha hocine Kerbaa , AMAR Mezache , Houcine OUDIRA , ,(2022), On the Performance of GLRT-LTD CFAR Processor in Correlated Pareto Clutter Under Different Estimators,2022 19th International Multi-Conference on Systems, Signals & Devices (SSD),Sétif, Algeria
- 2022
-
2022
Effect of Non-Integer Order Moments on Parameter Estimation of Pareto Distributed Clutter plus Noise
In this paper, accurate estimate of the Pareto plus noise shape parameter using a modified non-integer positive and negative order moment estimator is investigated. Closed form of the NIPNOME is derived in a previous work [16]. In single pulse case with a fixed value of the order moment, undesirable shape parameter estimates are obtained for high shape parameter values. In the prospect of improving the estimation performance, the NelderMead algorithm is used to optimize two parameters fitness function involving the shape and the non-integer order moment of the considered model. The impact of the order moment on the NIPNOME is examined firstly using simulated Pareto plus noise data. Then, via IPIX database, fitting comparisons are carried out using Pareto plus noise PDFs and CCDFs where [zlog(z)] and NIPNOME are employed to elucidate the robustness of the proposed estimator
Citation
Taha hocine Kerbaa , ,(2022), Effect of Non-Integer Order Moments on Parameter Estimation of Pareto Distributed Clutter plus Noise,2022 19th International Multi-Conference on Systems, Signals & Devices (SSD),Sétif, Algeria
- 2022
-
2022
Improved Decentralized SO-CFAR and GO-CFAR Detectors via Moth Flame Algorithm
Optimization of distributed constant false alarm rate (CFAR) system parameters is an essential part in radar detection applications. In this work, the moth flame algorithm (MFO) is proposed as an optimization tool to compute scale factors of distributed Greatest of-CFAR (GO- CFAR) and Smallest of-CFAR (SO-CFAR) detectors in presence of Gaussian disturbance. Local binary decisions are obtained firstly from different sensors, at the fusion center, a fusing rule is applied to obtain a global decision. Detection performances comparisons are conducted against previous works using Gray Wolf Optimization (GWO) and Biography Based Optimization (BBO) methods. Simulation results show that the proposed optimizer demonstrates a slight superiority in some cases for ensuring fixed probability of false alarm and higher detection probabilities.
Citation
Taha hocine Kerbaa , AMAR Mezache , Houcine OUDIRA , ADMIN Admin , ,(2022), Improved Decentralized SO-CFAR and GO-CFAR Detectors via Moth Flame Algorithm,2022 International Conference of Advanced Technology in Electronic and Electrical Engineering (ICATEEE),M'sila, Algeria
- 2020
-
2020
Parameter Estimation in Radar K-Clutter Plus Noise Based on Otsu’s Algorithm
In a previous work, it has been shown that the application of a modified fractional order moment (MFOM) estimator leads to the same accuracy as the [zlog(z)] method with lower computation complexity. However, undesirable estimation performances have been observed for single look data, low sample sizes and large values of the K-distribution shape parameter. Moreover, the application of positive and negative order moments estimators (PNOME) has a serious impact on the estimation accuracy of the shape parameter. To reduce this sensitivity, it is important to apply thresholding approaches in the case of a single pulse transmission. To this effect, single and double thresholding estimators are proposed in this paper and the Otsu’s algorithm is used to compute underlying thresholds. On the basis of Monte-Carlo simulation, the performances of the proposed estimators are assessed against moments and [zlog(z)] methods. Experiment examples indicate that the thresholding approaches based on the Otsu’s algorithm is more accurate with computational advantages than existing estimators.
Citation
Taha hocine Kerbaa , AMAR Mezache , Houcine OUDIRA , , (2020), Parameter Estimation in Radar K-Clutter Plus Noise Based on Otsu’s Algorithm, Ingénierie des systèmes d’ information, Vol:25, Issue:3, pages:295-302, IIETA
- 2020
-
2020
CNN-LSTM Based Approach for Parameter Estimation of K-Clutter Plus Noise
This paper concerns the problem of estimating the parameters of the K plus noise distribution. In a previous work, it has been shown that, in the multilook scenario, the modified fractional order moment estimator (MFOME) has about the same estimation accuracy as the [zlog(z)] method, but lower computational complexity. However, significant estimation errors have been observed in the single look scenario, low sample size, and large values of the K-distribution shape parameter. Moreover, the computational complexity of the [zlog(z)] estimator discourages its implementation in practical applications. The aim of this work is to estimate the shape parameter of the K-distribution with reduced computational complexity. The problem can be formulated as a supervised many-to-one sequence prediction. We propose here a hybrid model including convolutional and long-short-term-memory (LSTM) neural networks (NN). Estimation performance is investigated by processing both simulated and real clutter data.
Citation
Taha hocine Kerbaa , ,(2020), CNN-LSTM Based Approach for Parameter Estimation of K-Clutter Plus Noise,2020 IEEE Radar Conference (RadarConf20),Florence, Italy
- 2019
-
2019
Effect of Fractional Order Moments on Parameter Estimation of K-Clutter plus Noise
Parameter estimation of radar clutter is considered as a critical task for the development of target detectors. This work covers the shape parameter estimation of K-clutter plus noise using a modified fractional order moments based approach (MFOME). Closed form of the FOME with fixed fractional order moment is derived in a previous work [11] where undesirable estimation errors are produced in some cases with single look data and low sample sizes. In order to achieve better estimation performance, the fractional order moment and the shape parameter should be optimized together. To this effect, a numerical formula of the corresponding fitness function is given and unconstrained nonlinear optimization method based on the Nelder-Mead simplex algorithm is used to compute the unknown parameters. Via simulated K-clutter plus noise data, the effect of the fractional order on the estimation accuracy is studied firstly. Then, comparisons with existing HOME, FOME and [zlog(z)] methods are conducted to illustrate the efficiency of the proposed estimator.
Citation
Taha hocine Kerbaa , AMAR Mezache , Houcine OUDIRA , ,(2019), Effect of Fractional Order Moments on Parameter Estimation of K-Clutter plus Noise,2019 6th International Conference on Image and Signal Processing and their Applications (ISPA),Mostaganem, Algeria
- 2019
-
2019
Model Selection of Sea Clutter Using Cross Validation Method
This work concerns a model selection of sea radar clutter used for adaptive target detection. Three distributions without thermal noise are considered; K, Pareto type II and compound Gaussian inverse Gaussian (CG-IG) with scale and shape parameters. The model selection is carried out by comparing the experimental complementary cumulative distribution function (CCDF), drawn from the recorded data intensity, to a set of the CCDF curves derived from the underling models. To do this, the cross validation technique is used after dividing a set of data into four segments. The best distribution is selected in which the mean of the means square of errors (MSEs) between the real CCDF curve and the fitted CCDF curve is minimal. Fitting comparisons of models are illustrated through overall data of Intelligent PIxel X-band radar (IPIX). From this study, it is shown that the Pareto type II distribution is suited in several cases of a low cell resolution. On the other hand, the K and CG-IG models characterize generally sea clutter with medium and high cell resolutions.
Citation
Ahmed BENTOUMI , ADMIN Admin , Taha hocine Kerbaa , ADMIN Admin , AMAR Mezache , , (2019), Model Selection of Sea Clutter Using Cross Validation Method, Procedia Computer Science, Vol:158, Issue:, pages:Pages 394-400, Elsevier
- 2018
-
2018
Performance of Non-Parametric CFAR Detectors in Log-Normal and K Radar clutter
In this work, the performance of logt-, GMOS(Geometric Mean Order Statistic), TMOS(Trimmed MOS) and IE-CFAR (Inclusion/Exclusion) detectors are investigated in presence of log-normal and K distributed clutter. First, for a finite number of clutter samples, dependence of the false alarm probability P FA upon clutter parameters is examined. The CFAR property for the case of log-normal clutter is maintained while the P FA depends somewhat on the shape parameter of the K distribution. Then, by carrying out Monte-Carlo simulations, we show that for the case of log-normal clutter a small detection difference exists between the underlying CFAR detectors. In the case of K-distributed clutter, there is a significant detection difference for small values of the shape parameter (spiky clutter case).
Citation
Ahmed BENTOUMI , AMAR Mezache , Taha hocine Kerbaa , ,(2018), Performance of Non-Parametric CFAR Detectors in Log-Normal and K Radar clutter,2018 International Conference on Electrical Sciences and Technologies in Maghreb (CISTEM),Algiers, Algeria