Home // SENSORCOMM 2011, The Fifth International Conference on Sensor Technologies and Applications // View article

CPU-friendly Tracking in Wireless Sensor Networks

Authors:
Neeta Trivedi
Narayanaswamy Balakrishnan

Keywords: multitarget tracking, Bayesian filtering, sequential Monte Carlo, particle filter, sensor networks

Abstract:
The problem of tracking multiple targets using Sequential Monte Carlo technique under the framework of Bayesian techniques for wireless sensor networks is discussed. Distributed filtering in wireless sensor networks is an active area of research owing to the high communication costs of centralized tracking. However, distributed filtering must carefully address the conflict between high correlation among signals picked up by neighboring sensors and detached sensing by far away nodes due to limited sensing radii. Further challenges relate to the processing and communication of large number of particles in resource-constrained sensor nodes. This paper proposes a novel integrated approach to network management and target tracking by which distributed tracking is achieved in a lightweight manner. Important contributions are ‘Consensus Tracking’ as a low-cost distributed solution for sensor tasking and ‘Multitiling’ as computationally efficient solution for managing and propagating particles.

Pages: 261 to 267

Copyright: Copyright (c) IARIA, 2011

Publication date: August 21, 2011

Published in: conference

ISSN: 2308-4405

ISBN: 978-1-61208-144-1

Location: Nice/Saint Laurent du Var, France

Dates: from August 21, 2011 to August 27, 2011