CFAR Detection of Range-Spread Target in White Gaussian Noise Based on the Discrepancy of Distance of Successive HRRPs

摘要

Basing on high similarity of multiple successive high-resolution range profiles (HRRPs), a novel detector is proposed to improve detection performance, which utilizes the discrepant distance of HRRPs to detect range spread targets. Firstly, using the sliding cross correlation to eliminate the adverse effect of target speed. Secondly, Sigmoid-Shaped Nonlinear Shrinkage Map is employed to the pixel of every HRRPs to suppress noise. Finally, Minkowski distance is applied to construct the test statistic based on the discrepant distance of multiple successive HRRPs. The results of simulation experiments allow us to confirm that the proposed detector can achieve significant improvements compared with some existing detectors.