Detection of Random Targets in Sensor Networks with Applications

Authors:	Huang Minyi, University of Melbourne, Australia Nair Girish, University of Melbourne, Australia
Topic:	1.4 Stochastic Systems
Session:	Modeling and Estimation of Stochastic Systems
Keywords:	Hidden Markov models, sensor networks, optimal stopping, targetdetection, distributed change detection.

Abstract

This paper considers the detection of random targets modelled by ahidden Markov chain using distributed multiple sensors. Theoptimization problem is formulated as an optimal stopping problem whichintends to increase the capturing probability of the moving object.Finally, thegeneral formulation of this paper may be applied for distributed(change or failure) detection in a networked system,which generalizes the traditional quickest detection.