Title :
Robust DMR and multi-rate adaptive beamforming
Author :
Cox, Henry ; Pitre, Richard
Author_Institution :
ORINCON Corp., Arlington, VA, USA
Abstract :
Dominant mode rejection (DMR) is an approach to adaptive beamforming in which only the large eigenvalues of the correlation matrix and their eigenvectors are used. It requires fewer snapshots due to the reduced degrees-of-freedom. Once the partial eigendecomposition has been performed, a variety of opportunities arise to adjust the algorithms on a beam-by-beam basis without significant additional computations. A modification is developed to improve the robustness to signal mismatch. An approach to controlling the quiescent beam pattern that works for ocean acoustic noise is also presented. The standard approaches fail because source array elements are usually spaced closer than one-half wavelength so that the noise is correlated. Finally a two stage adaptive approach is introduced. It involves a slowly adapting algorithm optimized for nearly stationary conditions followed by a thresholded DMR that rapidly adapts to dynamic interference, but only when the interference is strong.
Keywords :
acoustic arrays; acoustic noise; acoustic signal processing; array signal processing; correlation methods; direction-of-arrival estimation; eigenvalues and eigenfunctions; interference (signal); matrix algebra; underwater sound; correlated noise; correlation matrix; degrees-of-freedom; dominant mode rejection; dynamic interference; eigendecomposition; eigenvalues; eigenvectors; minimum variance distortionless response; multi-rate adaptive beamforming; ocean acoustic noise; optimised algorithm; quiescent beam pattern; robust DMR; signal mismatch; snapshots; source array elements; stationary conditions; thresholded DMR; two stage adaptive approach; underwater acoustic applications; Acoustic beams; Acoustic noise; Array signal processing; Convergence; Eigenvalues and eigenfunctions; Frequency; Interference; Noise robustness; Subspace constraints; White noise;
Conference_Titel :
Signals, Systems & Computers, 1997. Conference Record of the Thirty-First Asilomar Conference on
Conference_Location :
Pacific Grove, CA, USA
Print_ISBN :
0-8186-8316-3
DOI :
10.1109/ACSSC.1997.680577