DocumentCode
2628021
Title
Collaborative Event Region Detection in Wireless Sensor Networks Using Markov Random Fields
Author
Wang, Tsang-Yi ; Yu, Chao-Tang
Author_Institution
Graduate Inst. of Commun. Eng., Nat. Chi Nan Univ.
fYear
2005
fDate
7-7 Sept. 2005
Firstpage
493
Lastpage
497
Abstract
Research on event region detection problems in wireless sensor networks (WSN) has received a great amount of interest recently. In these problems, sensor networks are asked to determine where the regions or boundaries in the environment. Each sensor could have its observation coming from its own hypothesis. Sensor nodes in WSN could have different true hypotheses. In this paper, the design of fusion rule when the sensor nodes observe different phenomenon is considered. Based on the Markov random fields (MRF) model and a new sensor fusion technique, we propose a MRF sensor fusion algorithm that can resolve the problem under consideration. The MRF is used to model the spatial correlation, and the sensor fusion technique considers the concept of sensor reliability. Some important issues in WSN, such as channel transmission errors and possible sensor faults are also considered in this paper. By numerical simulations, we have shown that the proposed MRF sensor fusion algorithm can result in a good detection performance
Keywords
Markov processes; numerical analysis; random processes; sensor fusion; telecommunication network reliability; wireless sensor networks; Markov random fields; channel transmission errors; collaborative event region detection; fusion rule; numerical simulations; sensor faults; sensor fusion technique; sensor nodes; sensor reliability; wireless sensor networks; Algorithm design and analysis; Collaboration; Event detection; Markov random fields; Monitoring; Numerical simulation; Sensor fusion; Sensor phenomena and characterization; Spatial resolution; Wireless sensor networks;
fLanguage
English
Publisher
ieee
Conference_Titel
Wireless Communication Systems, 2005. 2nd International Symposium on
Conference_Location
Siena
Print_ISBN
0-7803-9206-X
Type
conf
DOI
10.1109/ISWCS.2005.1547750
Filename
1547750
Link To Document