A least squares multiple constraint direct data domain approach for STAP

Traditionally, statistical space-time adaptive processing (STAP) approaches have been implemented to detect targets using an airborne radar system. These approaches typically assume a wide sense stationary environment and utilize auxiliary training data to estimate the statistics of the interference. Unfortunately the airborne radar environment can be highly non-stationary and may contain coherent interferers, which could drastically impact statistical STAP performance. Various approaches have been investigated to address this problem. One approach is the direct data domain least squares (D3LS) approach. This deterministic least squares adaptive signal processing technique operates on a "snapshot-by-snapshot" basis to determine the adaptive weights for nulling interferers and estimating signals of interest (SOI). Due to the reduced training data set (one range ring only-the range ring under test) and because of the particular least squares techniques employed, this approach can readily be implemented in real time in field deployable sensor signal processing systems. This paper presents an improvement to the existing D3LS approaches for radar processing. By implementing multiple space-time constraints system gain is maintained on the signal of interest when the signal arrives slightly off-set in angle, Doppler, or both. To illustrate this concepts an airborne radar simulation is used to generate radar data, and the D3LS processing results are presented.