Waveguide-Invariant-Based Ranging and Receiver Localization Using Tonal Sources of Opportunity

Acoustic emissions from cargo ships transiting coastal waterways, measured by a single hydrophone, can be exploited to estimate both the time-varying source-receiver range and receiver location. In this paper, two parameter-search-based maximum-likelihood estimators are presented: one for \beta, the waveguide invariant parameter, and one for source range. \beta characterizes the interference structure inherent to ducted acoustic propagation and is central to the spectral analysis involved in the range estimation. The source ranging method extends prior work by the authors and focuses on the strong narrowband tonals that typically dominate the acoustic spectra of cargo ships. Source ranging results are presented using real data from the SWellEx-96 experiment and are shown to be in close agreement with simulation results obtained using the KRAKEN normal mode program. A technique for estimating the position of a receiver, which can be stationary or moving, is also proposed. The localization technique requires knowledge of the source track, hydrophone data, and an initial estimate of the receiver's position. An application to autonomous underwater vehicle (AUV) navigation is included in which the waveguide-invariant-based range estimates are used to mitigate the position estimation error due to drift in inertial measurement units on submerged AUVs. A specific scenario is examined, using SWellEx-96 data, in which an initial position error of 3 km is reduced to under 1 km using the proposed technique.

Duke Scholars

Author Jeffrey L. Krolik Electrical and Computer Engineering