An Embedded Real-time Passive Underwater Acoustic Localization System using a Compact Sensor Array

Abstract

In this thesis, a passive underwater acoustic localization system using a compact sensor array is developed which receives underwater acoustic sensor data and outputs the position estimations to a map display in real-time. Through simulation, it is evaluated using 130 kHz pulses, which is representative of the harbour porpoise echolocation clicks. The localization system is tested in several environments including the Aquatron, Bay of Fundy, Herring Cove, and New Zealand. The Time Difference of Arrival localization algorithm is used to estimate the position of sound sources using the difference of propagation time between multiple sensors. The implementation also improves upon a traditional grid search by using a lookup table stored in a hyperoctree to reduce the execution time of a position estimation. Additionally, a method to analyze and reduce the estimation error for different sensor geometries is developed. Finally, the impact of noise is mitigated by using various pre-processing techniques.