Saving Energy in Wireless Sensor Networks Based on Echo State Networks

Authors:
Ling Qin;Rongqiang Hu;Qi Zhang
Affiliations:
Department of Information Engineering, Wuhan University of Technology, Wuhan, P.R. China 430070;Department of Information Engineering, Wuhan University of Technology, Wuhan, P.R. China 430070;The National Key Laboratory of EMC, Wuhan, P.R. China 430064
Venue:
ISNN 2009 Proceedings of the 6th International Symposium on Neural Networks: Advances in Neural Networks - Part III
Year:
2009

Citing 7
Cited 0

Distributed Detection and Data Fusion

Distributed Detection and Data Fusion
Dynamic Power Management in Wireless Sensor Networks

IEEE Design & Test
TEEN: ARouting Protocol for Enhanced Efficiency in Wireless Sensor Networks

IPDPS '01 Proceedings of the 15th International Parallel & Distributed Processing Symposium
Energy-Efficient Communication Protocol for Wireless Microsensor Networks

HICSS '00 Proceedings of the 33rd Hawaii International Conference on System Sciences-Volume 8 - Volume 8
Energy-Efficient Link Layer for Wireless Microsensor Networks

WVLSI '01 Proceedings of the IEEE Computer Society Workshop on VLSI 2001
Energy efficient system partitioning for distributed wireless sensor networks

ICASSP '01 Proceedings of the Acoustics, Speech, and Signal Processing, 200. on IEEE International Conference - Volume 02
An application-specific protocol architecture for wireless microsensor networks

IEEE Transactions on Wireless Communications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Prolonged lifetime, robustness and scalability were important requirements in Wireless Sensor Networks (WSNs). We investigated the problem of energy-saving in Wireless Sensor Networks using Echo State Networks (ESN). In this research field, one key factor of the problems was how to save energy efficiently in battery-driven sensor nodes. We tried to present an approach addressing these difficulties based on ESN learning information of these sensors' history status when only the data was available. Echo State Networks utilized incremental updates driven by new sensor readings and massive short memory with history inputs, thus varying communication rates can help save energy. We evaluated this method against those traditional approaches to save energy, and observe that the quality of the overall operation was comparable to the approaches. Therefore, the ability of Echo State Networks to prolong lifetime during the sensor network operation made this approach more suitable and applicable.