Fuzzy statistical normalization for target detection in active sensing data

A fuzzy statistical normalization for target detection in active sensing data is proposed in this paper. The first stage of the fuzzy statistical normalization is the Fuzzification of the active sensing data. Then the fuzzy inference and defuzzification operation based on statistical method and alpha-cut approach are performed, which not only attenuate the heavier tailed clutter data values, but also enlarge the lower shadow area noise data values. The constant false alarm rate (CFAR) detector based on fuzzy statistical normalization firstly estimates the background level with the normalized data, and then detects the target signal with the original active sensing data based on the estimated background level. Performance comparison between the proposed CFAR detector with outlier rejection based on fuzzy statistical normalization and the conventional CFAR detectors is carried out to validate the superiority of the proposed fuzzy statistical normalization in CFAR detection. The results show that the CFAR detector with fuzzy statistical normalization is robust.