Target-depth estimation in active sonar

International audienceIn active sonar, the objectives are to detect, localize and classify an underwater target. Azimuth and range are often used in anti-submarine warfare to localize targets. The depth may also be used as the key tactical information for strategy purposes or as a good feature for target classification or discrimination. Two dimensional arrays as ank arrays, cylindrical arrays, and hullmounted arrays have access to elevation angles. Even linear towed arrays can give some information about the elevation using the different conical bearings measured when multipath propagation arises. In the context of long ranges and summer Mediterranean sound-speed profile (SSP), this paper presents a new target-depth estimation method, which uses elevation and arrival time measures from one sonar ping in a multipath environment. This method is based on ray back-propagation with a probabilistic approach. This localization algorithm minimizes the mean-squared error of elevation angles at the receiver and arrival times between a model and measures. This method is tested through Monte-Carlo simulations of classic active sonar scenarios and using experimental data from a real reduced-scaled tank. In active sonar, acoustic waves can take the same path on the way back or another path, so ray path combinations can occur. Our localization method discusses also about this ray identification, or how these combined acoustic paths were managed