A new distributed constant false alarm rate detector

Performance of distributed CFAR test under various clutter amplitudes

“…5). If the homogeneous background noise power in all the sensors are nearly identical, then the MOS test provides a much better performance than the OR rule (and the NTS test) [13]. Since the performance of biased NTS is similar to that of the unbiased NTS, we only consider the biased NTS (called NTS for short) for the remainder of the performance analysis.…”

“…If a is close to 1, then the MOS test performs much better than the Boolean OR and the Boolean AND fusion rules [13].…”

“…If we assume that¸0 i is the same for all i = 1,2,:::, n, then the MOS test defined below is a CFAR test [13]:…”

“…In this test, the sum of the test samples is compared with an adaptive threshold obtained by the product of a fixed scaling factor and the maximum of the received order statistics, to decide the presence/absence of a target. It was shown in [13] that MOS provides a considerable performance gain over OR or AND fusion rules. In deriving the above test, the problem formulation assumes that the test cells of different sensors all have statistically identical noise (clutter), and that if a target is present in the surveillance regions, all the test cells have statistically identical target returns.…”

“…A review of different distributed CFAR target detection techniques is given in [12]. In [13], a new distributed CFAR test called the maximum order statistic (MOS) test was proposed. In this test, the sum of the test samples is compared with an adaptive threshold obtained by the product of a fixed scaling factor and the maximum of the received order statistics, to decide the presence/absence of a target.…”