Distributed detection and tracking in sensor networks
- 31 October 2003
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 1174-1178 vol.2
- https://doi.org/10.1109/acssc.2002.1196968
Abstract
Our work develops a framework for using a wireless ad-hoc network of distributed sensors for detection and tracking of moving objects. Possible applications might be vehicular traffic monitoring in highways or in battlefields, detecting and tracking people movement in secure areas, clots flowing through the bloodstream or even nuclear devices moving around cities. The central idea is to utilize decentralized processing within the network to build a better picture of what is going on (locally), before reporting the results (centrally). This notion, which relies on collaboration among sensors, reduces false alarms and avoids sending all the sensor measurements back to a centralized processing unit. Simple assumptions are made about the structure of the network and the capabilities of the individual nodes. Uncertainty is introduced in the form of uncertain node positions, uncertain position estimates, false events and an unknown number of objects.Keywords
This publication has 7 references indexed in Scilit:
- Collaborative classification applications in sensor networksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Instrumenting the world with wireless sensor networksPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Detection, classification, and tracking of targetsIEEE Signal Processing Magazine, 2002
- Wireless integrated network sensorsCommunications of the ACM, 2000
- Blind beamforming on a randomly distributed sensor array systemIEEE Journal on Selected Areas in Communications, 1998
- Data fusion and tracking using HMMs in a distributed sensor networkIEEE Transactions on Aerospace and Electronic Systems, 1997
- An algorithm for tracking multiple targetsIEEE Transactions on Automatic Control, 1979