Abstract

This thesis demonstrates the application of information geometry to problems in underwater acoustics. Information geometry combines the fields of information theory and differential geometry by interpreting a multiparameter model as a Riemannian manifold in an ambient data space. Information geometry tools are especially powerful in context of problems of experimental design and model selection in ocean acoustics. The application area specifically considered in this research is geoacoustic inversion, where seabed parameter values are inferred from acoustical data. Chapter 2 contains a paper submitted to the Journal of Theoretical and Computational Acoustics, which introduces information geometry tools such as the model manifold and Fisher information in context of a review of work in underwater acoustics doing parameter sensitivity analysis. An example constructing model manifolds for a sound propagation model is given in the second half of the paper. Chapter 3 contains a paper submitted to the Journal of the Acoustical Society of America, Express Letters, which constructs model manifolds for a sound propagation model and compares the information content of absolute and relative transmission loss in regards to seabed parameters, demonstrating how information geometry can be use to inform experimental design. This thesis contains the initial application of information geometry to ocean acoustics, with many more advances that can be pursued in future work.

Degree

MS

College and Department

Computational, Mathematical, and Physical Sciences; Physics and Astronomy

Rights

https://lib.byu.edu/about/copyright/

Date Submitted

2024-07-25

Document Type

Thesis

Keywords

information geometry, underwater acoustics, model manifold, sound propagation, Fisher information, sensitivity analysis, Cramer-Rao bounds, geoacoustic inversion, reduced-order modeling, optimal experimental design, transmission loss

Language

english

Download

Share

COinS