Diver Detection and Localization Using Passive Sonar

An open-circuit scuba diver’s acoustic signature (radiated sound) consists predominately of a sequence of regularly spaced wideband pulses each corresponding to the inhaling phase of the diver’s breathing cycle. A cyclic frequency analysis of the output signal from a single hydrophone leads to the automated detection of the diver and an estimate of the diver’s breathing rate. Measurement of the differential time of arrival (DTOA) of the radiated sound at a pair of widely separated hydrophones requires computation of the wideband cross-ambiguity function. By fitting a DTOA model to the measurements for pairs of sensors over a sufficiently long period of time enables estimation of the diver’s motion parameters. Results are presented for real data collected in a shallow water experiment where an opencircuit scuba diver swam at constant speed and altitude above and along the axis of a horizontal linear array which consisted of eight hydrophones uniformly spaced at 14 m and located 1 m above the sea floor.