Performance of a new greatest of selection CFAR detector based on order statistics and censored mean

A new CFAR detector (OSCMGO) based on order statistics (OS) and censored mean (CM) is introduced. It uses the greatest value of the OS and CM local estimations as a noise power estimator to set an adaptive threshold, and it also takes into account the automatic censoring technique proposed by He You(see IEE Proc.-Radar, Sonar Navig., vol.141, no.8, p.205-212,1994). Under Swerling II assumption, the analytic expressions of P_fa, P_d and ADT of OSCMGO are derived. By comparison with other schemes, the results show that the detection performance of OSCMGO is acceptable in a homogeneous background, and it is more robust in the presence of multiple interfering targets, while the sample sorting time of OSCMGO is less than half of that of OS. Since the greatest of selection logic is adopted, OSCMGO will be more effective to control the false alarm rate at clutter edges