Multiband FIR digital filter design algorithm for radar clutter suppression

An algorithm is presented for designing multi-band FIR digital filters to suppress wideband or bimodal clutter in MTI radars. The technique is an iterative process in which the passband edges are varied to maximize the usable bandwidth over which a signal is detected with acceptable probability while maintaining specified clutter rejection. The clutter-bandwidth tradeoff is shown to be a function of such interrelated filter parameters as number of taps, transition widths, passband ripple, variable stopband attenuation, and stopband width. A 25-tap bandstop filter design is presented which provides 70% usable bandwidth while maintaining 20 dB wideband clutter rejection. The design is then modified to include an additional 50 dB ground clutter attenuation. The resulting 26-tap multi-band filter meets all specifications while still retaining 60% usable bandwidth. An alternative technique for bimodal clutter rejection is presented which utilizes the natural null at dc for an N-even bandstop filter to attenuate the ground component.