Brian G. Ferguson
Defence Sci. and Technol. Organisation, P. O. Box 44, Pyrmont, NSW 2009, Australia
Both conventional and adaptive wave-vector filtering techniques are applied to microphone data from an experimental land-based acoustic surveillance array and to hydrophone data from a horizontal line array towed below the sea surface. After transformation from the time domain to the frequency domain, the outputs of each array's acoustic sensors are processed in the spatial domain by each of the wave-vector filtering techniques. A comparison of the results shows that the data-adaptive technique, which requires inversion of the cross-power spectral matrix, offers superior performance by suppressing spatial leakage in the wave-number domain, enhancing the spatial resolution through narrower beamwidths and providing superdirective array gain at frequencies well below the spatial Nyquist frequency of the array. Adaptive wave-vector filtering is shown to be a powerful diagnostic tool for analyzing the self-noise in the towed array.