Abstract
The deployment of infrastructure in the ocean is often slowed due to the potential harm of animals in the area. In the Bay of Fundy, the deployment of tidal farms have been explored for several years, however there is a large population of harbour porpoises. In this work, a remote passive localization system is proposed to track harbour porpoises using the sound emitted by the animal. In comparison to active sonar systems, the proposed sensor does not send a pulse. This reduces the potential noise pollution, as well as the energy consumption of the system. In order to implement this real-time system, two core components are developed: the localization algorithm, as well as the communication link to return the information to the surface. A complete design procedure is described in this report, including the analytical development, the modelling and the testing of the system. The proposed real-time algorithm relies on a compact hydrophone array that is designed to detect bearing and range. The algorithm complexity is optimized to be programmed on a field programmable gate array (FPGA); the FPGA is used for parallel computing of the acoustic data. The algorithm is tested using calibration data in realistic conditions, as well as with real harbour porpoise echolocation samples that were collected in the Bay of Fundy in the Summer 2019. The communication system as well as the hardware for the localization system are deployed in the Aquatron. The analytical models, validated by the measurement results demonstrate the great potential for magneto-inductive communication as a means to cross the air-water interface. Additionally, these tests were used to confirm how the sensor array can be connected to the processor to enable real-time localization. Overall, the developed platform has great potential as a smart remote sensor node that can be used to monitor the ocean in real-time.