Multidimensional localization of multiple sound sources using frequency domain ICA and an extended state coherence transform

Recently, direction-of-arrival (DOA) and position estimation for acoustic signals has been studied intensively and many different algorithms have been proposed. Among different solutions for multiple sources, blind source separation (BSS) methods have drawn much attention due to their good performance. In this paper, we present a localization algorithm using the results from a frequency domain independent component analysis (ICA) algorithm combined with an extended version of the state coherence transform (SCT). We motivate the SCT as an approximated maximum likelihood (ML) approach and compare our localization algorithm with the steeredresponse power with phase transform (SRP-PHAT) and the averaged directivity pattern (BSS-ADP) algorithm. 2D localization results show superior performance of our algorithm.

[1]  Michael S. Brandstein,et al.  Robust Localization in Reverberant Rooms , 2001, Microphone Arrays.

[2]  S. Araki,et al.  Multiple source localization using independent component analysis , 2005, 2005 IEEE Antennas and Propagation Society International Symposium.

[3]  Scott C. Douglas,et al.  Scaled Natural Gradient Algorithms for Instantaneous and Convolutive Blind Source Separation , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[4]  W. Kellermann,et al.  A Real-Time Demonstrator for the 2D Localization of two Sound Sources Using Blind Adaptive MIMO System Identification , 2008, 2008 Hands-Free Speech Communication and Microphone Arrays.

[5]  Bin Yang,et al.  Disambiguation of TDOA Estimation for Multiple Sources in Reverberant Environments , 2008, IEEE Transactions on Audio, Speech, and Language Processing.

[6]  Walter Kellermann,et al.  Multidimensional localization of multiple sound sources using averaged directivity patterns of Blind Source Separation systems , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.