Learning-Aided Geoacoustic Inversion

Abstract

Geoacoustic inversion (GI) methods aim at inferring the underwater environment, including bottom properties, from acoustic signals received by one or multiple hydrophones. This project aims to combine trans-dimensional (TransD) inversion with particle flow to reduce processing time for large-scale GI problems. In addition, we will use invertible neural networks (INNs) and advanced methods forsparsity-promoting estimation in the context of GI. An important aspect of the proposed research will be the evaluation of geoacoustic model parameter uncertainties. The developed methods will be used to analyze GI data collected from the seabed characterization experiment 2017 (SBCEX-17) and the shallow water mid-frequency experiments 2021 (SWMFEx-21) and 2022 (SWMFEx-22).This abstract is approved for public release.

Document Details

Document Type
DoD Grant Award
Publication Date
Nov 21, 2023
Source ID
N000142412021

Entities

People

  • Florian Meyer

Organizations

  • Office of Naval Research
  • United States Navy
  • University of California, San Diego

Tags

Readers

  • Acoustical Oceanography.
  • Distributed Systems and Data Platform Development

Technology Areas

  • AI & ML