Biological versus Subspace Methods in Sound Localization.

摘要

Sound localization is determining the location of sound sources using the measurements of the signals received by an array of sensors. Humans and animals possess the natural ability of localizing sound. Researchers have tried to model nature's way of solving this problem and have come up with different methods based on various neuro- physiological studies. Such methods are called biological methods. On the other hand, there is another community of researchers who has looked at this problem from pure signal processing point of view. Among the more popular methods for solving this problem using signal processing techniques are the subspace methods. In this thesis, a comparative study is done between biological methods and subspace methods. Further, an attempt has been made to incorporate the notion of head-related transfer function in the modeling of subspace methods. The implementation of a biological localization algorithm on a DSP board is also presented.