Enhanced Minimum Variance Beamforming

摘要

It is shown that both the multiple emitter signal location wavenumber spectrum estimator (MUSIC) and the minimum variance distortionless response (MVDR) beamformer are just special cases of the enhanced minimum variance (EMV) beamformer. Moreover, by separating planewave arrivals into two types, i.e., high level interferences and low (threshold) level signals, it is illustrated how the EMV beamformer simultaneously gives MUSIC-like angular resolution of an interference and MVDR-like detection of threshold signals. An additional advantage of the EMV beamformer is that the number of threshold signals need not be known a priori. As a preprocessor for the EMV beamformer, a dimensionality reducing transformation of the array data vector to a subarray beam space is proposed. This transformation is wavefront invariant and expedites the separation of source and coherent noise subspace eigenvalues. Keywords: Beamforming; Cross spectral density matrix; Eigenstructure; Interfering source; Minimum variance; Threshold signal. (jhd)