A Low-Complexity Method of Signal Subspace Fitting

A low-complexity method for signal subspace fitting is proposed based on the mu lti-stage Wiener filter(MSWF) can span a constringent signal subspace,which is completely equivalent to t he Krylov subspace.Secondly,a new criterion function for signal subspace fittin g is derived,and then a novel method for direction of arrival (DOA) estimation is developed.Analysis indicates that the proposed method can work very well even i n the case where the rank of the MSWF is much lower than the number of signals.S ince finding the constringent signal subspace merely involves the forward recurs ion of the MSWF,does not need to estimate the array covariance matrix or comput e its eigenvectors,the proposed method is computationally efficient.Finally,the effectiveness of the proposed approach is verified by numerical results.