Master Thesis: Localization of killer whale calls

Be part of an interdisciplinary team and contribute to the signal processing methods we develop to analyze the communication of wild killer whales. The goal is a publication about the new method and its application to our own recordings.

Keywords: bioacoustics, dolphin, phase locked loop, array signal processing, direction of arrival, source separation

Research context
To date, the best model systems for understanding language evolution have been non-human primates and songbirds. However, cetaceans represent an evolutionary peak in terms of their cognitive capacities, complex communication systems and their multilevel societies, providing striking parallels to some human societies. Cetaceans display a behavioral skill set that make them an ideal study system for questions surrounding the evolution of complex communication.

Our approach
We follow a data driven approach and develop state of the art recording technology to acquire large datasets of underwater acoustic signals of communicating cetaceans. The challenge is to separate all calls from the individual animals within a group. We address this challenge with nested short and long base-line hydrophone arrays and signal processing methods to separate and localize the individual calls. Isolated calls serve the study of phonetics and vocal repertoire whereas the resolved group communication is the foundation for exploring the linguistic structure of cetacean communication systems.

Master thesis
We aim to extend our recently developed algorithm for pitch tracking with phase locked loops (PLL), from the single hydrophone case to an array of four hydrophones arranged in a tetrahedron. This will allow us to estimate the direction of arrival for every call and to better separate overlapping calls.

Outline
As an outline of the project we propose: code and literature review, project plan, implement a model to simulate signals, prepare an example dataset, adapt the pitch tracking algorithm to multiple hydrophones, analyze performance, contribute to the scientific publication. We have weekly meetings to discuss outcomes, ideas and next steps. The thesis workload is designed for 6-month full-time work.

Your benefits
You will acquire proficiency in bioacoustics, a rapidly growing field for ecological monitoring and in audio signal processing valuable for divers careers in technology and sciences. Possibly you can participate in the next expedition and operate your algorithm in real time.

Your profile
We are looking for a motivated student with interests in bioacoustics, signal processing and programming. Depending on the interests of the applicant we can tailor the scope of the independent work. To apply please send a CV and transcript of records to one of the contacts

Collaborators:
Dr. Jörg Rychen: jrychen (at) ethz.ch
Dr. Kaje Wierucka: kaja.wierucka (at) uzh.ch
Supervisor:
Prof. Dr. Richard Hahnloser: rich (at) ini.ethz.ch

© 2024 Institut für Neuroinformatik