Fuzzy soft decision CFAR detector for the K distribution data

A fuzzy statistical normalization fuzzy constant false alarm rate (FSNF-CFAR) detector in a K distribution background based on fuzzy statistical normalization and fuzzy soft decision is proposed. The performance of the proposed fuzzy soft decision detector is studied both for homogeneous backgrounds and for nonhomogeneous environments caused by interfering targets or clutter edges. Performance comparisons with conventional hard decision CFAR detectors such as cell averaging CFAR (CA-CFAR), greater of CFAR (GO-CFAR), and ordered statistics CFAR (OS-CFAR) are carried out. The simulation shows that the proposed FSNF-CFAR detector is simple and efficient, and the comparison results show that it not only can get good detection performance in homogeneous K distribution backgrounds but also can confront interfering targets and clutter edges at the same time in nonhomogeneous environments. Moreover, the fuzzy soft decision detector can provide more valuable information than the hard decision detector for data fusion, target tracking, or object identification.