Performance Analysis of Linearly Combined Order Statistics CFAR Processors in Heterogeneous Background

In this article, we present an exact performance analysis of linearly combined order statistics (LCOS) constant false alarm rate (CFAR) processors in heterogeneous background. We provide detailed derivations of the false alarm probability (
$P_\mathrm{fa}$
) and detection probability (
$P_\mathrm{d}$
) for LCOS-CFAR processors in a completely heterogeneous background. Then, we consider a two-class case, where the reference cells consist of independent random variables from one of two nonidentical exponential populations, to obtain as simple expressions for the
$P_\mathrm{d}$
in the multiple target scenario and
$P_\mathrm{fa}$
in the clutter edge as possible. These analytical expressions are validated by Monte Carlo simulations. Utilizing these expressions, we compare the performance of several LCOS-CFAR processors in multiple-target scenarios and clutter edge.