Token based algorithm for supporting mutual exclusion in opportunistic networks

  • Authors:

  • Sagar A. Tamhane;Mohan Kumar

  • Affiliations:

  • University of Texas at Arlington, TX;University of Texas at Arlington, TX

  • Venue:
  • MobiOpp '10 Proceedings of the Second International Workshop on Mobile Opportunistic Networking
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Opportunistic networks are essentially distributed networks with transient connectivity among nodes. Nodes in opportunistic networks are resource constrained, mobile and infrequently come in contact with each other. In such a distributed network, nodes may require exclusive access to a shared object or resource. Ensuring freedom from starvation is a challenging problem in opportunistic networks due to the infrequent contacts and node failures. In this paper, we review mutual exclusion algorithms proposed for generic mobile ad hoc networks (MANETs) and discuss their applicability to opportunistic networks. Further, we propose a novel token based algorithm and prove it is free from starvation and deadlock, and satisfies the safety property. Unlike existing algorithms for MANETs, the proposed algorithm does not require continuous monitoring of the network topology. Simulation results show that our algorithm is communication efficient as compared to other algorithms proposed for generic mobile ad hoc networks. We also propose a timeout based fault detection algorithm that exploits the inter contact time distributions. To the best knowledge of the authors, the proposed algorithm is the first to support mutual exclusion in opportunistic networks.