Share Email Print
cover

Proceedings Paper

Self-localization of wireless sensor networks using self-organizing maps
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Recently there has been a renewed interest in the notion of deploying large numbers of networked sensors for applications ranging from environmental monitoring to surveillance. In a typical scenario a number of sensors are distributed in a region of interest. Each sensor is equipped with sensing, processing and communication capabilities. The information gathered from the sensors can be used to detect, track and classify objects of interest. For a number of locations the sensors location is crucial in interpreting the data collected from those sensors. Scalability requirements dictate sensor nodes that are inexpensive devices without a dedicated localization hardware such as GPS. Therefore the network has to rely on information collected within the network to self-localize. In the literature a number of algorithms has been proposed for network localization which uses measurements informative of range, angle, proximity between nodes. Recent work by Patwari and Hero relies on sensor data without explicit range estimates. The assumption is that the correlation structure in the data is a monotone function of the intersensor distances. In this paper we propose a new method based on unsupervised learning techniques to extract location information from the sensor data itself. We consider a grid consisting of virtual nodes and try to fit grid in the actual sensor network data using the method of self organizing maps. Then known sensor network geometry can be used to rotate and scale the grid to a global coordinate system. Finally, we illustrate how the virtual nodes location information can be used to track a target.

Paper Details

Date Published: 28 March 2005
PDF: 8 pages
Proc. SPIE 5803, Intelligent Computing: Theory and Applications III, (28 March 2005); doi: 10.1117/12.606812
Show Author Affiliations
Emre Ertin, The Ohio State Univ. (United States)
Kevin L. Priddy, Jacobs Sverdrup (United States)


Published in SPIE Proceedings Vol. 5803:
Intelligent Computing: Theory and Applications III
Kevin L. Priddy, Editor(s)

© SPIE. Terms of Use
Back to Top