Target localization in a reverberant shallow ocean waveguide with environmental uncertainty using a nonlinear frequency-difference signal processing technique

Abstract

Model-based signal processing for active sonar localization is typically infeasible due to insufficient knowledge of the acoustic environment. Additionally, in a shallow ocean environment, surface or bottom roughness create diffuse reverberation that can often obscure a desired target echo. Recently, a nonlinear signal processing technique was developed for passive acoustic source localization (Worthmann, et al., 2015) that is applicable to high frequency sources in uncertain shallow ocean environments. This technique exploits a nonlinear field product (the autoproduct) and uses in-band hydrophone array measurements to determine field information in a lower, out-of-band, frequency regime where environmental uncertainties are less detrimental. When extended to monostatic active sonar with a vertical array, this technique allows a model-based signal processing algorithm to combat the detrimental effects of reverberation. The nonlinear signal processing algorithm is presented, along with simulations in a 5-km range, 200-m deep ideal waveguide with environmental uncertainties and significant reverberation at frequencies between 2- and 5-kHz. Successful detection and localization of a mid-water-column target is found to be possible at simulated signal-to-reverberation levels as low as −5 dB. Comparisons to existing signal processing detection and localization algorithms are provided. [Sponsored by the Office of Naval Research and the National Science Foundation.]