Classification of distant targets situated near channel bottoms

Identification algorithms are considered for a class of targets situated near the bottom of a water channel. It is assumed that the target-sensor distance relative to the channel depth is such that a ray-based representation of the scattered fields is appropriate (vis-à-vis a modal representation). Two approaches are considered for processing the scattered fields. In one algorithm a deconvolution is performed to remove the channel response, and thereby recover the free-field target scattered signature. In this case the classifier is trained based on free-field data. In the second approach the array receiver is employed to point the sensor in particular directions, and the beam-formed signal is used directly in the subsequent classifier. In this case the classifier must be trained based on in-channel data. Multiple scattered signals are measured, from a sequence of target-sensor orientations, with the waveforms classified via a hidden Markov model. Example results are presented for scattering data simulated via the finite-element method and coupled to a normal-mode waveguide modal, for elastic targets situated in a water channel.