Scholarlink

The Pareto Type II distribution has proven to be a robust statistical model for capturing the characteristics of high-resolution sea clutter, particularly under low grazing angle conditions. The application of the Generalized Likelihood Ratio Test Linear Threshold Detector (GLRT-LTD) within such clutter environments necessitates precise estimation of clutter parameters to maintain optimal detection performance. In this study, we conduct a comprehensive evaluation of several parameter estimation techniques-namely Maximum Likelihood Estimation (MLE), integer-order moments (HOME), fractional-order moments (FOME), and the \(z \log z\) estimator-under conditions of correlated Pareto-distributed clutter. Through extensive Monte Carlo simulations, we demonstrate that while MLE and \(z \log z \) estimators yield comparably accurate results in general, the \(z \log z\) method provides superior performance in scenarios involving strong correlation among clutter samples. This highlights the estimator’s robustness and computational efficiency, particularly in challenging detection environments where correlation effects are non-negligible.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Performance of GLRT-LTD CFAR Processor in Correlated Pareto Clutter

  • 章节

  • 发表日期：

    2026年

查看全文
  • Taha Hocine Kerbaa,
  • Amar Mezache,
  • Houcine Oudira

摘要

The Pareto Type II distribution has proven to be a robust statistical model for capturing the characteristics of high-resolution sea clutter, particularly under low grazing angle conditions. The application of the Generalized Likelihood Ratio Test Linear Threshold Detector (GLRT-LTD) within such clutter environments necessitates precise estimation of clutter parameters to maintain optimal detection performance. In this study, we conduct a comprehensive evaluation of several parameter estimation techniques-namely Maximum Likelihood Estimation (MLE), integer-order moments (HOME), fractional-order moments (FOME), and the \(z \log z\) estimator-under conditions of correlated Pareto-distributed clutter. Through extensive Monte Carlo simulations, we demonstrate that while MLE and \(z \log z \) estimators yield comparably accurate results in general, the \(z \log z\) method provides superior performance in scenarios involving strong correlation among clutter samples. This highlights the estimator’s robustness and computational efficiency, particularly in challenging detection environments where correlation effects are non-negligible.