ZOULIKHA Douiou
ضويو زليخة
zoulikha.douiou@univ-msila.dz
0658045007
- Departement of ELECTRONICS
- Faculty of Technology
- Grade PHd
About Me
Science et Technologies
Filiere
Télécommunications
Location
Msila, Msila
Msila, ALGERIA
Code RFIDE- 1999-01-11 00:00:00
-
ZOULIKHA Douiou birthday
- 2025-12-07
-
2025-12-07
Robust Moment-Based Estimators and Optimization Algorithms for Parameter Extraction in Compound Gaussian Radar Clutter with Log-Normal Texture under Thermal Noise
Accurate parameter estimation in radar systems affected by Compound Gaussian clutter with Log- Normal Texture (CG-LNT) and thermal noise is addressed through an integrated approach that merges moment-based techniques with numerical optimization. The framework incorporates Higher-Order Moment Estimator (HOME), Fractional Order Moment Estimator (FOME), Fractional Negative Order Moment Estimator (FNOME), and logarithmic-moment-based methods, enhanced by the Nelder–Mead simplex algorithm to improve precision in noisy environments. Evaluation relies on both Monte Carlo– generated synthetic clutter and real high-resolution radar measurements. Performance is quantified using Mean Square Error (MSE) and statistical fit analyses via Probability Density Functions (PDFs) and Complementary Cumulative Distribution Functions (CCDFs). Findings reveal that combining moment-based estimators with optimization algorithms yields notable improvements in estimation accuracy, especially under high clutter variability, offering a robust solution for advanced radar detection and CFAR operations in non-Gaussian, noise-contaminated conditions.
Citation
zoulikha douiou , ,(2025-12-07), Robust Moment-Based Estimators and Optimization Algorithms for Parameter Extraction in Compound Gaussian Radar Clutter with Log-Normal Texture under Thermal Noise,National Conference on Electrical Engineering NCEE’2025,Djillali Liabès University of Sidi Bel-Abbes Electrical Engineering Faculty
- 2025-11-23
-
2025-11-23
Study in Compound-Gaussian Lognormal Texture Modeling
Accurate estimation of clutter parameters plays a crucial role in the performance of modern radar detectors, particularly under heavy-tailed non-Gaussian environments. This paper examines the robustness and statistical consistency of moment-based estimators applied to the Compound-Gaussian model with Lognormal Texture (CG-LNT). The considered estimators include the Higher Order Moment (HOME), Fractional Order Moment (FOME), Zlog(z), and Fractional Negative Order Moment (FNOME) techniques. Through extensive Monte Carlo simulations, the sensitivity of each estimator to the texture variability and fractional order selection is analyzed. Results highlight that the FNOME maintains stable estimation accuracy across a broad range of clutter spikiness, outperforming other methods in terms of mean-squared error (MSE) and bias behavior. These findings emphasize the suitability of fractional negative-order formulations for reliable modeling of non-Gaussian sea clutter in high-resolution radar applications.
Citation
zoulikha douiou , ,(2025-11-23), Study in Compound-Gaussian Lognormal Texture Modeling,The Fibonacci International Conference on Engineering, Technology, and Mathematics,Rome, Italy.
- 2025-08-23
-
2025-08-23
Performance Analysis of the FNOME Estimator under Thermal Noise for CG-LNT Radar Clutter
Parameter estimation of radar clutter modeled by the Compound-Gaussian with Lognormal Texture (CG-LNT) distribution is crucial for improving detection performance in high-resolution radar systems. The CG-LNT model, a heavy-tailed and non-Gaussian distribution, is widely used to characterize sea clutter variations. This paper revisits the Fractional Negative Order Moment Estimator (FNOME) and evaluates its robustness under different levels of additive thermal noise. Through extensive Monte Carlo simulations, we analyze the estimation accuracy of the FNOME for the texture mean δ and standard deviation σ, using the Mean Squared Error (MSE) as the performance metric. The results confirm that FNOME maintains high estimation accuracy even in the presence of noise, demonstrating its suitability for real-world radar environments where clutter is non-Gaussian and noise is non-negligible. This contribution provides a deeper understanding of FNOME's capabilities beyond the noise-free case and validates its application in thermally disturbed radar clutter.
Citation
zoulikha douiou , ,(2025-08-23), Performance Analysis of the FNOME Estimator under Thermal Noise for CG-LNT Radar Clutter,8th International Conference on Applied Engineering and Natural Sciences ICAENS 2025,Konya,Turkey
- 2025-05-07
-
2025-05-07
On the Parameter Estimation of CG-LNT Radar Clutter
Abstract Parameter estimation of radar clutter modeled by the Compound-Gaussian with Lognormal Texture (CG-LNT) distribution is crucial for improving detection performance in high-resolution radar systems. The CG-LNT model, a heavy-tailed and non-Gaussian distribution, is widely used to characterize sea clutter variations. In paper, comparative study of the performance of the different methods of estimation is conducted. The estimators considered are higher order moment estimator (HOME), fractional order moment estimator (FOME), [zlog(z)] estimator and fractional negative order moment estimator. Furthermore, the impact of the fractional order in the FOME and FNOME estimator is also considered in his paper, where the performance of these two estimators is tested using different values of the fractional order. Through Monte Carlo simulations, the estimation accuracy of the different estimators is evaluated for both the mean and the standard deviation of the CG-LNT distribution. The Mean Squared Error (MSE) as the performance metric.
Citation
zoulikha douiou , ,(2025-05-07), On the Parameter Estimation of CG-LNT Radar Clutter,the first national conference on renewable energies and advanced electrical engineering NC-REAEE'2025,Msila
- 2024-07-09
-
2024-07-09
Fractional Negative Order Moment Parameter Estimator of Compound-Gaussian Clutter with Lognormal Texture
parameter estimation of Compound-Gaussian with lognormal texture (CG-LNT) distribution is considered. The CG-LNT distribution is nonGaussian heavy-tailed distribution, it used in radar domain to describe the variation of high-resolution sea clutter. The probability density function (PDF) is characterized by two parameters, the standard deviation and the mean. This paper proposes a closed form estimator of CG-LNT distribution parameters based on fractional negative order moment (FNOME). Comparative study is established against the existing estimators existing in the literature to evaluate the efficiency of the proposed estimator using CG-LNT simulated and real sea clutter data, the mean square error (MSE) criterion is also used to measure the estimation accuracy.
Citation
zoulikha douiou , IZZEDDINE Chalabi , ABDELKADER Djerad , , (2024-07-09), Fractional Negative Order Moment Parameter Estimator of Compound-Gaussian Clutter with Lognormal Texture, PRZEGLĄD ELEKTROTECHNICZNY, Vol:100, Issue:7, pages:42-45, SEP – Polish Electrical Engineers Association (SEP)
- 2023-11-22
-
2023-11-22
Performance analysis of parameter estimation for compound Gaussian with log-normal texture clutter in the presence and absence of thermal noise
Abstract— Accurate parameter estimation plays a pivotal role in enhancing radar detection performance. This paper addresses parameter estimation for the CG-LNT (Compound Gaussian with Log-Normal Texture) distribution. The study meticulously examines both situations involving the presence and absence of thermal noise, comparing the performance of estimation methods, HOME (Higher-Order Moments Estimation) , FOME (Fractional Order Moment Estimation), [zlog(z)], and curve fitting, using real radar data. The estimation performance is evaluated by the MSE (Mean Squared Error criterion). The results show the ability of the CG-LNT in presence of thermal noise to model high- resolution sea clutter.
Citation
zoulikha douiou , ,(2023-11-22), Performance analysis of parameter estimation for compound Gaussian with log-normal texture clutter in the presence and absence of thermal noise,2ETA,university mohamed el bachir el ibrahimi of bordj bou arreridj