Adaptable misbehavior detection and isolation in wireless ad hoc networks using policies

  • Authors:
  • Oscar F. Gonzalez Duque;Antonis M. Hadjiantonis;George Pavlou;Michael Howarth

  • Affiliations:
  • Centre for Communications Systems Research, University of Surrey, Guildford, UK;Centre for Communications Systems Research, University of Surrey, Guildford, UK;Networks and Services Research Lab, University College London, UK;Centre for Communications Systems Research, University of Surrey, Guildford, UK

  • Venue:
  • IM'09 Proceedings of the 11th IFIP/IEEE international conference on Symposium on Integrated Network Management
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Wireless ad hoc networks provide the communications platform for new technologies and applications, such as vehicular ad hoc networks or wireless mesh networks. However, their multihop wireless nature makes them inherently unreliable and vulnerable, since their overall performance depends on the cooperative packet forwarding behavior of each individual node. In this paper we present a role-based approach that uses a distributed management overlay and gathers information about the packet forwarding activities of each node in the network. Using policies to control an adaptive algorithmic method that monitors the individual behavior of each node, we show that it is possible to detect, accuse and punish misbehaving nodes with a high degree of confidence. Our evaluation results demonstrate that after the successful detection of misbehaving nodes, their punishment through network isolation can significantly improve network performance in terms of packet delivery and throughput.