Evaluation of Sound Source Localization Algorithms with Recorded Binaural Signals

The localization of binaural sound sources is essential for various modern applications. While most algorithms struggle with dynamic situations caused by the movement of the head, the movement invariant beamforming binaural direction of arrival estimator (MIBF DoA estimator) involves the information on head rotations. This algorithm applies the motion information to a target beamforming direction-of-arrival (DoA) estimator in order to identify the position of a sound source on the horizontal plane in dynamic situations.

This thesis implements an extension of this algorithms to three dimensions, achieved by the usage of quaternions for the MIBF DoA estimator. We evaluate the algorithm based on simulations and data from an experimental set up. The results demonstrate that the algorithm identifies sound source positions in the three dimensional space correctly, especially during
the movement of the head around multiple rotation axes. Furthermore, the utilization of head motions resolves problems caused by an ambiguous interaural time difference (ITD) and interaural level difference (ILD).

All interested parties are cordially invited, registration is not required.

General information on the colloquium, as well as a current list of dates of the Communication Technology Colloquium can be found at:
https://www.iks.rwth-aachen.de/aktuelles/kolloquium

Anett Schindler 
Institute of Communication Systems(IKS)
Prof. Dr.-Ing. Peter Jax
RWTH Aachen University
Muffeter Weg 3a, 52074 Aachen, Germany
+49 241 80 26958(phone)
+49 241 80 22254(fax) 
schindler@iks.rwth-aachen.de
https://www.iks.rwth-aachen.de/