Date of Award

Fall 2021

Publication Type

Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

First Advisor

J. Ahamed

Second Advisor

M. Hassanzadeh

Third Advisor

S. Erfani

Keywords

Acoustic signal processing, Localization, Modal propagation, Underwater source localization

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

Acoustic localization plays a pivotal role in underwater vehicle systems and marine mammal detection. Previous efforts adopt synchronized arrays of sensors to extract some features like direction of arrival (DOA) or time of flight (TOF) from the received signal. However, installing and synchronizing several hydrophones over a large area is costly and challenging. To tackle this problem, we use a single-hydrophone localization system which relies on acoustic signal processing methods rather than multiple hydrophones. This system takes modal dispersion into consideration and estimates the distance between sound source and receiver (range) based on dispersion curves. It is shown that the larger the range is, the more separable the modes are. To make the modes more distinguishable, a non-linear signal processing technique, called warping, is utilized.

Propagation model of low-frequency signals, such as dolphin sound, is well-studied in shallow water environment (depth D<200 m), and it was demonstrated that at large ranges (range r>1 km), modal dispersion is utterly visible at time frequency (TF) domain. We used Peker is model for the aforementioned situation to localize both synthetic and real underwater acoustic signals. The accuracy of the localization system is examined with various sounds, including impulsive signal, sounds with known Fourier transform, and signals with estimated source phase. Experimental results show that the warping technique can considerably lessen the localization error, especially when prior knowledge about the source signal and waveguide are available.

Share

COinS