6. Performance of GLRT-LTD CFAR Processor in Correlated Pareto Clutter

This chapter investigates the performance of the Generalized Likelihood Ratio Test with Linear Threshold Detector (GLRT-LTD) Constant False Alarm Rate (CFAR) processor in environments characterized by correlated Pareto clutter. The study focuses on the challenges of radar detection in complex scenarios, particularly in maritime surveillance and low-grazing-angle operations where clutter and noise can obscure target signals. The chapter explores various parameter estimation methods, including the Higher Order Moments Estimator (HOME), Fractional Order Moments Estimator (FOME), a closed-form estimator, and the Maximum Likelihood Estimator (MLE), and their impact on the detection performance of the GLRT-LTD algorithm. Through extensive Monte Carlo simulations, the chapter demonstrates that the accuracy of parameter estimation is strongly influenced by the degree of correlation in the clutter. In low-correlation scenarios, FOME and MLE perform reliably across different shape parameter regimes. However, in high-correlation environments, the closed-form estimator outperforms all others, offering a compelling trade-off between computational efficiency and robustness. The chapter also confirms that the GLRT-LTD detector preserves CFAR properties, even when operating with adaptively estimated parameters. Detection performance analysis further validates the effectiveness of the closed-form estimator, particularly in scenarios with severe clutter correlation. The findings position the closed-form estimator as a strong candidate for integration into real-time radar detection systems, offering a robust and efficient alternative to traditional MLE-based frameworks. The chapter concludes with potential directions for future research, including the integration of these estimators within more complex radar system architectures and their extension to other heavy-tailed models and Bayesian inference-based detectors.