A time-domain backpropagating ray technique for source localization

A new method is presented for source localization which is based on a ray backpropagation procedure, and which exploits both the temporal and spatial characteristics of the multipath arrival structure at a receiving sensor array. The basic method consists of two steps. The first step employs a receiving sensor array to estimate the angles and relative arrival times of the various multipath trajectories which arrive at the receiver after having been emitted by the source located at an unknown range R and depth Z from the receiver. The second step utilizes a ray tracing technique to backpropagate the rays and produce an estimate of the source position by making full use of all the spatial and temporal information extracted by the array. The source signal is taken to be a narrow-band stochastic process, which simulates a realistic situation commonly encountered in the ocean acoustic environment. Some key advantages of the approach are (1) localization ambiguity is minimized by utilizing ray travel times as well as path convergence; (2) the approach does not require full-wave matched-field processing; (3) ray backpropagation is numerically very efficient; (4) phase ambiguity is eliminated by employing the group delay of arriving wave packets; (5) ray chaos can be avoided simply by excluding chaotic rays from the backpropagation algorithm.