Robust Waveform Design for Multistatic Cognitive Radars

In this paper, we propose robust waveform techniques for multistatic cognitive radars in a signal-dependent clutter environment. In cognitive radar design, certain second order statistics such as the covariance matrix of the clutter, are assumed to be known. However, exact knowledge of the clutter parameters is difficult to obtain in practical scenarios. Hence, we consider the case of waveform design in the presence of uncertainty on the knowledge of the clutter environment and propose both worst-case and probabilistic robust waveform design techniques. Initially, we tested our multistatic, signal-dependent model against existing worst-case and probabilistic methods. These methods appeared to be over conservative and generic for the considered scenario. We therefore derived a new approach where we assume uncertainty directly on the radar cross-section and Doppler parameters of the clutters. Accordingly, we propose a clutter-specific stochastic optimization that, by using Taylor series approximations, is able to determine robust waveforms with specific signal to interference and noise ratio outage constraints.