Title: Wireless sensor network for multi-target detection algorithm based on blind source separation
Authors: Pengju He; Xiaomeng Chen; Huazhi Zeng
Addresses: Department of Automation, Northwestern Polytechnical University, Xi'an 710129, China ' Department of Automation, Northwestern Polytechnical University, Xi'an 710129, China ' Department of Automation, Northwestern Polytechnical University, Xi'an 710129, China
Abstract: Detecting multi-target using wireless sensor network is a research hotspot in recent years, and it has broad application prospect. Aiming at the problem that signal mixture exists in wireless sensor networks for multi-target detection, this paper proposed a blind source separation (BSS) algorithm to process mixed signal. The algorithm took detected multi-channel signal as research object. The number of sources in signals was determined utilising eigenvalue method firstly, transforming over-determined BSS into well-posed. Then separation of signals was achieved using non-negative matrix factorisation (NMF). Experiments utilised four channel man-made signals and two speeches separately to validate the proposed algorithm. Results showed presented algorithm could determine the number of sources in multi-channel signals and it can be used to solve signal mixture problem in multi-target detection.
Keywords: wireless sensor networks; WSNs; multi-target detection; BSS; blind source separation; NMF; non-negative matrix factorisation; mixed signals; multi-channel signals.
International Journal of Security and Networks, 2016 Vol.11 No.4, pp.235 - 241
Received: 11 Aug 2015
Accepted: 19 Aug 2015
Published online: 26 Sep 2016 *