A rare event approach to the detection of target-like signals in CFAR training data

A new method based on the existence of rare events (RE) is proposed to detect the presence of nonhomogeneous samples in a set of Constant False Alarm Rate (CFAR) training data. Two RE schemes designated as the mean-to-mean ratio (MMR) and the variance-to-variance ratio (VVR) tests are proposed. No a priori knowledge of the nonhomogeneity topology is assumed. Analysis using Monte-Carlo method based on Rayleigh clutter and Swerling I target models is presented. Target-like interferences which seriously degrade the detection performance of the cell-averaging CFAR detector can be detected with a higher probability by RE detectors.