RADAR CLUTTER SUPPRESSION USING NOTCH FILTER BANK IN MULTIBEAM RADAR

A new clutter suppression method is proposed for a multibeam radar with which the profile of the input signal can be estimated with a statistically high accuracy. In this method, the suppression filter is selected adaptively from the discrete notch filter bank on the basis of the estimated clutter bandwidth. As the selection method, the concept of selection probability linked to the bandwidth estimation accuracy is introduced so that the selection algorithm is derived. To evaluate the performance of the system based on this algorithm, the conventional improvement factor is improved by defining the performance function taking into consideration the filter selection probability and the desired detection probability. Based on these factors, a configuration example of a notch filter bank is presented for the model radar. The system performance is evaluated by the figure of merit (FOM) defined here. The results obtained are: (1) there exists an optimum maximizing the FOM in the clutter bandwidth estimation accuracy and the filter distribution spacing; and (2) if the estimation accuracy is increased in response to the distribution spacing of the filter bank, the FOM is improved as the distribution spacing is reduced.