Decentralized Reactive Clustering for Collaborative Processing in Sensor Networks

  • Authors:
  • Yingyue Xu;Hairong Qi

  • Affiliations:
  • The University of Tennessee, Knoxville, TN;The University of Tennessee, Knoxville, TN

  • Venue:
  • ICPADS '04 Proceedings of the Parallel and Distributed Systems, Tenth International Conference
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

A sensor network forms a loosely-coupled distributedenvironment where collaborative processing among multiplesensor nodes is essentail in order to compensate forthe limitation of each sensor node in its processing capability,sensing capability, and energy usage, as well asto improve the degree of fault tolerance.Due to the sheeramount of nodes deployed, collaboration is usually carriedout among nodes within the same cluster.Differentclustering protocols can affact the performance of networkto a great extent.Most existing clustering protocolseither do not adequately address the energy-constraintproblem or derive clusters proactively which may not besuitable for event-driven collaborative processing in sensornetworks.This paper focuses on the design of clusteringprotocols for collaborative processing.We proposea decentralized reactive clustering (DRC) protocol wherethe clustering procedure is initial only when events aredetected.It uses power control technique to minimizeenergy usage in forming clusters.We compare the performanceof DRC with another popular clustering algorithm,LEACH.Simulation results show considerable improvementover LEACH in energy conservation and networklifetime using DRC.