Stochastic Models and Robust Estimation for Broadband Acoustic Mode Signals

Funded by an Office of Naval Research Ocean Acoustics Entry-Level Faculty Award
Dates:   April 2002 - September 2005

Abstract
The goal of this project is to develop statistical models and robust estimation techniques for broadband mode signals propagating in fluctuating ocean environments. In deep water the low-order modes constitute some of the most energetic arrivals at long ranges. Analysis of both simulated and experimental data indicates that, for ranges on the order of megameters, these arrivals have a complicated structure due to internal-wave-induced coupling. Understanding the mechanisms and effects of mode coupling is a prerequisite for using the mode signals in applications such as tomography or matched field processing. Building on the results of the Acoustic Thermometry of Ocean Climate Experiment, this research synthesizes three topics in underwater acoustics: environmental characterization, signal processing, and propagation modeling. The specific objectives of the proposed project are: to characterize mode wavenumber and shape fluctuations associated with temperature changes; to develop robust signal processing techniques for broadband mode estimation; and to investigate stochastic models of mode coupling due to internal waves. Analysis of environmental and acoustic data collected during the North Pacific Acoustic Laboratory experiment is an integral part of this research.