Interarray noise correlation
Evan F. Berkman, Jude R. Nitsche · The Journal of the Acoustical Society of America · 1978
Small amounts of residual correlated noise may degrade the tracking performance of split arrays with high values of processing gain. In the absence of directional interference, it is usually assumed that the noise output of subarrays are uncorrelated if the subarray acoustic centers are far apart relative to an acoustic wavelength. This assumption is usually justified by the fact that the correlation length for an isotropic noise field is on the order of an acoustic wavelength. However, the correlation length of the noise field after being spatially filtered by each subarray is more nearly on the order of a subarray effective aperture. Thus, one wonders if subarray noise output will have small correlation coefficients only if the acoustic centers are very well separated relative to a subarray effective aperture. For usual split array configurations, the distance between acoustic centers is equal to or only slightly greater than the effective subarray aperture. Hence, typically the interarray noise correlation is determined by the close-in side lobe structure of the interarray correlation function. We find that corresponding sidelobe levels of the interarray correlation function are usually much smaller than those of the omniphone correlation function so that small interarray noise coherency may be anticipated, even though the array acoustic centers are only slightly further apart than the characteristic correlation length of the interarray noise correlation function. The interarray noise correlation function is shown to be convolution of a) the incident noise field or omniphone correlation function and b) the auto-correlation of the subarray shading function. The implications of this relationship are examined.