Automatic Censoring Best Linear Unbiased Constant False Censoring and Alarm Rates Detector in Lognormal Background

Abstract

We deal with the problem of automatic censoring with constant false censoring rate (CFCR) and automatic target detection with constant false alarm rate (CFAR), i.e., CFCAR procedures, against non-stationary Lognormal clutter with unknown scale and shape parameters. We show that, in multiple target situations, automatic censoring detectors outperform those based on fixed point censoring. For that, we study the censoring and detection performances of the Automatic Censoring Best Linear Unbiased (ACBLU) CFCAR detector. The censoring and detection algorithms are a two in one built detector. They select repeatedly a suitable set of the ranked reference cells to estimate the unknown background level and set the adaptive threshold accordingly. To this end, the Lognormal probability density function (pdf) is reduced to a Normal one, via a logarithmic transformation, and the parameters are estimated using the best linear unbiased estimators (BLUEs). Performances of this detector are carried out using Monte Carlo simulations.