Large aperture sonar arrays have small resolution cells which (i) limit the number of "snapshots" available for accumulating sample covariance matrices and (ii) resolve the multipath propagation between each source and the array. Moving ships with high bearing rates transit resolution cells quickly setting limits upon short term stationarity assumptions. Usually, this leads to "snapshot deficient" processing which must be "patched" with techniques such as diagonal loading or subspace methods. Moreover, in most sonar environments there are several paths connecting a source to a receiver. The resulting interference pattern can be resolved with large aperture sonar arrays operated off "broadside". We examine the problems associated with large aperture sonar arrays here, particularly in the context of nulling multiple strong discretes from moving ships.
[1]
I. Reed,et al.
Rapid Convergence Rate in Adaptive Arrays
,
1974,
IEEE Transactions on Aerospace and Electronic Systems.
[2]
N. R. Goodman,et al.
Probability distributions for estimators of the frequency-wavenumber spectrum
,
1970
.
[3]
Christ D. Richmond.
PDF's, confidence regions, and relevant statistics for a class of sample covariance-based array processors
,
1996,
IEEE Trans. Signal Process..
[4]
Henry Cox,et al.
Robust adaptive beamforming
,
2005,
IEEE Trans. Acoust. Speech Signal Process..
[5]
W. Kuperman,et al.
Matched field processing: source localization in correlated noise as an optimum parameter estimation problem
,
1988
.