Fast recursive subspace updating for detection and angle estimation of multiple sources
Kai‐Bor Yu · 2003
This paper is concerned with detection of a number of sources and estimation of their corresponding direction-of-arrival using eigenvalue decomposition updating with subspace structure constraint. The covariance matrix is a low rank matrix plus a noise diagonal matrix. In addition to cleaning up the data, the subspace constraint enables the author to develop an efficient algorithm for updating the principal subspace and the noise eigenvalue.>