Adaptive detection of distributed targets in noise and interference which is partially related with targets

Abstract

This paper studies adaptive detection of distributed targets in interference and noise. The signal and interference are assumed to lie in two known subspaces which are partially related; this denotes the cases where the signal and interference cannot be completely separated in spatial, temporal and frequency domains. Since impartible, the signal and interference are recast as one with the aid of singular value decomposition (SVD); then, the generalized likelihood ratio test (GLRT) and the two-step GLRT are derived in both homogeneous and partially homogeneous environments. The four new detectors are the generalizations of the existing GLRT-based ones; they have the constant false alarm rate (CFAR) properties and the capabilities of interference rejection. The effectiveness of the new detectors is demonstrated via numerical experiments, also in comparison with previous detectors of similar kind.