Target localization in range and depth from a passive linear towed array

Stealthily estimating the range and depth of a sound source from a submarine passive sonar is of capital tactical importance: it allows to adopt the appropriate escape or attack strategy towards the threat. The main difficulty in passive localization is that the transmitted waveform is unknown. Nevertheless, array processing can reveal some features of the acoustic field allowing localization, such as the arrival angles of the multiple paths, and the time delays between them. In this paper, we have compared two methods. The first one is a direct method, which simulates the field features from a set of possible source locations, and compares them to the measured features. The second one, an indirect method, uses the measured features to backpropagate the acoustic field. The former method has already been tested on simulated and real data in (Bennaceur et al., in OCEANS’18 MTS/IEEE, Charleston) and (Bennaceur et al., in OCEANS 2021, Singapore-U.S. Gulf Coast). Here, both methods are applied on real signals from a linear towed array, using bearings only and both bearings and time delays. The main advantage of these methods is that they are automatic in the sense that no sonar operator is required to estimate the source location.