A Novel Constant False Alarm Rate Detector Based on Ordered Data Variability
Zhong Xiao-feng · Torpedo Technology · 2009
In order to make full use of reference cell information and decrease constant false alarm rate(CFAR) loss,a new CFAR detector(named MOSODV-CFAR) based on ordered statistics(OS) and ordered data variability(ODV) is proposed.Its leading window and lagging window use OS method and ODV method to create two local noise estimations respectively,then the sum of the two estimations is taken as a global noise power estimation.Assuming type of target is Swerling II and noise has Gaussian distribution,an analytic expression of false alarm rate for MOSODV under homogeneous background is derived.In Comparison with mean ordered statistics MOSCM and MOSAC methods,simulation results show that the MOSODV is endowed with better detection performance under homogeneous environment and multi-target interference,and it exhibits better performance of false alarm control against clutter edge situations.