MLE-based order statistic automatic CFCAR detection in Weibull background

Abstract

In this paper, we address the problem of automatic target detection in Weibull clutter and multiple target situations, without any prior knowledge of neither the non stationary clutter statistics in which the radar operates nor the number of outliers that may be present in the reference window. In doing this, we develop the Forward and Backward Order Statistic Automatic Constant False Censoring and Alarm Rates Detectors based upon the Maximum Likelihood Estimator, MLE-based F/B-OSACDC-FCAR. The censuring and detection algorithms are a two in one built detector. They select repeatedly a suitable set of ranked cells among all reference cells surrounding the cell under test to estimate the unknown background level and set the adaptive threshold accordingly. The performance of these detectors is evaluated by means of Monte Carlo simulations.