Distributed labelling of audio sources in wireless acoustic sensor networks using consensus and matching

• Published in: 2016 24th European Signal Processing Conference (EUSIPCO) pp. 2345 - 2349
• Main Authors: Bahari, Mohamad Hasan, Plata-Chaves, Jorge, Bertrand, Alexander, Moonen, Marc
• Format: Conference Proceeding
• Language: English
• Published: EURASIP 01.08.2016
• Subjects: Clustering algorithms; consensus and matching; Distributed labelling; energy signatures; Estimation; Labeling; Microphones; non-negative independent component analysis; Signal processing algorithms; wireless acoustic sensor networks; Wireless communication; Wireless sensor networks
• ISSN: 2076-1465
• DOI: 10.1109/EUSIPCO.2016.7760668

In this paper, we propose a new method for distributed labelling of audio sources in wireless acoustic sensor networks (WASNs). We consider WASNs comprising of nodes equipped with multiple microphones observing signals transmitted by multiple sources. An important step toward a cooperation between the nodes, e.g. for a voice-activity-detection, is a network-wide consensus on the source labelling such that all nodes assign the same unique label to each source. In this paper, a hierarchical approach is applied such that first a network clustering algorithm is performed and then in each sub-network, the energy signatures of the sources are estimated using a non-negative independent component analysis over the energy patterns observed by the different nodes. Finally the source labels are obtained by an iterative consensus and matching algorithm, which compares and matches the energy signatures estimated in different sub-networks. The experimental results show the effectiveness of the proposed method.