A new CFAR processor in nonhomogeneous background and multiple target situations

The authors consider the problem of CFAR (constant false-alarm rate) detection in nonhomogeneous clutter and in multiple target situations. The data discriminator (DD) CFAR detector which performs two passes over the data, is proposed and analyzed. In the first pass, the algorithm senses possible interfering targets, and in the second pass the algorithm selects those samples which are identically distributed with the noise in the test cell to form the detection threshold. The false alarm control and the detection performance of the DD-CFAR processor are evaluated and compared to those of the ACGO and TM-CFAR detectors for nonhomogeneous background environments. In addition, the required processing time of the proposed detector is shown to be less than the required processing time of both the ACGO and TM-CFAR detectors