Abstract
In a random internal wave field where temporal varying water-column soundspeed profiles are very difficult to measure, matched-field source localization methods can be greatly degraded due to water-column model mismatch. In this paper, the data nullspace projection is applied to this source localization technique to reduce the effect of unknown water-column soundspeed variations. The basis of this method is to project the acoustic signal onto a data nullspace that is insensitive to water-column soundspeed fluctuations. This method only requires the mean and the second-order statistics of temporal varying water-column soundspeed profiles to calculate soundspeed empirical-orthogonal-functions. It does not require measurements of the exact soundspeed field, i.e., each snapshot, for the matched-field processing. In a simulation test case, a linear wave model is used to generate a random internal wave field, an acoustic source continuously transmits a single frequency tone, and the matched-field processing is implemented with the signal received on both a VLA and a HLA, respectively, for localizing the source position. The simulation results show that applying the data nullspace projection method can dramatically improve the robustness and accuracy of the matched-field source localization, resulting in a random internal wave field.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call