Acoustic self-localization of a wireless sensor network

• Published in: The Journal of the Acoustical Society of America 2005-04, Vol.117 (4_Supplement), p.2610-2610
• Main Authors: Schmidt, Peter L., Williams, Stephen M., Frampton, Kenneth D.
• Format: Article
• Language: English
• ISSN: 0001-4966; 1520-8524
• DOI: 10.1121/1.4809382

One of the biggest challenges to the field of wireless sensor networks is self-localization: that is the determination of the relative and absolute coordinates of each sensor node in the network. Previous work has been done to locate hydrophone arrays. However sensor networks have some unique constraints that make this more challenging. A typical application would involve the distribution of hundreds or thousands of sensor nodes over an area either by hand, airdrop or other means. One of the primary constraints on such a system is that centralized processing of self-localization data may be prohibitively complex. Furthermore, the data may be incomplete, contain reflected path events, and may be subject to other mission specific constraints. Therefore, a distributed computational scheme has been developed to solve acoustic time-of-arrival equations. A priori information about some sensor locations and user triggered source localization events are used along with a regularized inversion solution. Results of this will be presented based on both small scale experiments and larger systems simulations. Limits of the types of a priori information required for accurate results are detailed, along with studies of the accuracies obtained using various distributed calculations parameters.