A singular value decomposition-based method for solving a deterministic adaptive problem

A computational scheme based on the singular value decomposition (SVD) for a deterministic, data domain approach to the adaptive processing problem is presented. In the direct data domain approach, a single snapshot is consid ered for an assumed direction of arrival with unknown amplitude. This unkno wn signal strength is estimated on a snapshot by snapshot basis. The new SV D based method is compared with the QZ method for determining the generaliz ed eigenvalues of a system. Their performance in estimating the strength of signals of interest in the presence of main beam jammers, clutter, and the rmal noise is considered. Limited examples have been presented to illustrat e the two methods. (C) 1999 Academic Press.