Reduced-Rank STAP for Target Detection in Heterogeneous Environments

In an airborne radar context, heterogeneous situations are a serious concern for space-time adaptive processing (STAP), where the required secondary training data have to be target free and homogeneous with the tested data. Consequently, the performance of these detectors is severely impacted when facing a heavily heterogeneous environment. Single data-set algorithms such as the maximum likelihood estimation detector (MLED) algorithm, based on the amplitude and phase estimation (APES) method, have proved their efficiency in overcoming this problem by only working on primary data. However, restricting the estimation domain solely to the primary data often implies an inaccurate estimation of the covariance matrix. In this paper, we demonstrate that we can use reduced-rank STAP on the single data-set APES method to increase the performance of the STAP processing. We also introduce an algorithm that reduces the computational cost of the standard subspace-based algorithms based on eigenvalue decomposition. The results on realistic data show that reduced-rank methods outperform traditional single data-set methods in detection and in clutter rejection.