New Non-Stationary Target Feature Detection Techniques

In a seminal paper, two algorithmic versions of the multichannel parametric adaptive matched filter (PAMF) applied to space-time adaptive processing (STAP) in an airborne radar application were shown to achieve superior test detection statistics over the conventional adaptive matched filter (AMF), which uses a non-parametric approach to estimate the detection weight vector. In fact, the performance of the PAMF approach is very close to the ideal matched filter (MF) detection statistics under exactly known covariance (the clairvoyant case). Improved versions of the two original multichannel PAMF algorithms, one new multi-channel PAMF algorithm, and a new two-dimensional PAMF algorithm (all four with fast computational implementations) have been summarized in recent papers. In this paper, we provide the detection performance of the four improved/new PAMF algorithms with simulated radar data. In all cases, the performance is at least comparable to, and in some cases superior to, the original multi-channel PAMF algorithms presented by M. Rangaswany et al ( 2000), while achieving computational savings over the originals.