Threshold optimization for distributed CFAR detection in Weibull clutter using genetic algorithms

The use of genetic algorithms (GAs) tool for the solution of constant false alarm rate (CFAR) detection for Weibull clutter statistics is considered. An approximate expression of the probability of detection (P_D) of the ordered statistics (OS)-CFAR detector in Weibull clutter is derived. Optimal threshold values of distributed maximum likelihood (ML)-CFAR detector and distributed OS-CFAR detector with a known shape parameter of the background statistics are obtained using GAs. For the ML-CFAR distributed detection, we consider also the case when the shape parameter is unknown. A performance assessment is carried out, and results are given as a function of the shape parameter of the Weibull distribution and of the system parameters.