Parallel Matlab Computation for STAP Clutter Scattering Function Estimation and Moving Target Estimation

Estimation and Moving Target Estimation * Roger Chamberlain, Daniel R. Fuhrmann, John Maschmeyer, and Lisandro Boggio Washington University, St. Louis, Missouri Phone: 314-935-5708 Email Address: [email protected] Email Address: [email protected] Abstract Conventional moving target estimation in STAP radar applications is based, in part, on adaptive clutter scattering function estimation techniques. These techniques classically rely on radar return data from adjacent range gates to estimate the clutter scattering function for the range gate of interest. Here, we are interested in using geographical information systems in conjunction with accurate platform positioning information as the basis for the clutter scattering function estimation. The goal is to improve the effectiveness of moving target estimation techniques by providing additional information to the decisionmaking process. Our problem formulation is consistent with the model for space-time adaptive processing (STAP) presented in [1]. A pulse-Doppler radar platform with multiple transmit/receive elements emits several pulse train along an arbitrary flight path, such as a circle around the region of interest. Each pulse train is assumed to be perfectly coherent within one coherent processing interval (CPI), but different pulse trains are assumed noncoherent with respect to one another. The ground region is subdivided into pixels, or ground patches. The range and angle of each ground patch with respect to the platform for each transmitted pulse is assumed known, along with the illumination pattern. The received data for one pulse is modeled as the sum of the returns from all of the ground patches, each modulated by the transmit illumination. The data from all pulses or viewpoints is modeled in this way. Maximum-likelihood methodology is used to estimate the unknown scattering function [2] and a variant of the Adaptive Matched Filter [3] is used for moving target estimation.