Abstract:

Aiming at the difference existing at each statistical range cells’ output signal to clutter ratio, the rangespread target detection method based on range cell selecting is investigated. Firstly, the generalized likelihood ratio test based on order statistics (OS-GLRT) is analysed, then two kinds of modified range cell selecting detectors based on OS-GLRT are proposed. The first one adopts the rule of OR and fuses the decision results of all OS-GLRT to make a final decision, where the OS-GLRT includes all assumptions about the number of valid range cells. The modified OS-GLRT does not need any information about the number of valid range cells of rangespread targets. The second one is based on dual threshold (DH) technique, which uses a fixed primitive threshold to estimate the number of valid range cells and a dynamic secondary threshold to judge. Compared with the common detectors, the novel detectors need not know any prior information on rangespread targets and they are much more robust.