A Matched Mode Processing Method for the Depth Discrimination of a Moving Source

The mode distribution can be expressed as the wavenumber spectrum distribution, providing a basis for the matched mode processing to discriminate the source depth through depth estimation. The synthetic aperture formed by a moving source can be used to beamform the wavenumber spectrum, saving the physical aperture. However, the high-resolution spectrum also requires the source to move a distance of kilometers, so as to ensure a reliable depth estimate. We relax the source depth estimation into the source depth discrimination in this paper, based on the distinction in mode distribution between surface and submerged sources on the wavenumber spectrum. The characteristics that surface sources cannot excite low-order modes is still discernable on the wavenumber spectrum, when the aperture is too small to resolve all the modes. In this method, a small-aperture synthetic beamforming is applied to get a low-resolution mode distribution on the wavenumber domain. The spectral peaks are matched with the calculated mode functions to yield the depth ambiguity function, from the peak of which one can discriminate the source depth, despite the peak may not represent a reliable depth estimate. The needed aperture will be greatly reduced in this method. The feasibility is illustrated by a numerical example, where the aperture for the 49Hz source depth discrimination is required only a quarter of the one for source depth estimation.