Morphological Component Analysis and STAP Filters

STAP is an adaptive filtering algorithm that works on space-time radar data to suppress the effects of clutter and jamming and achieve both target identification and parameter estimation in airborne and space based radar [1][2]. The signals to which STAP processes are applied consist of three basic parts, target, clutter and jamming. In morphological component analysis (MCA) [3] the component signals are separated by assuming that each signal component can be sparsely represented in some suitable dictionary. It is assumed that expansions of a given signal component in the dictionaries of the other signal components are non-sparse. The MCA algorithm seeks a solution through an optimization method that implements this sparse-representation idea. This approach was applied to the analysis of ISAR images in [4]. In this paper we consider the basic STAP equations as a problem in multi-channel morphological component analysis (MMCA) [5] [6] [7]. The method is illustrated by application to the KASSPERI datacube.