Ad Hoc Networks. First International Conference, ADHOCNETS 2009, Niagara Falls, Ontario, Canada, September 22-25, 2009. Revised Selected Papers

Research Article

Distributed Detection of Wormhole Attacks in Wireless Sensor Networks

Download99 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-11723-7_14,
        author={Rennie Graaf and Islam Hegazy and Jeffrey Horton and Reihaneh Safavi-Naini},
        title={Distributed Detection of Wormhole Attacks in Wireless Sensor Networks},
        proceedings={Ad Hoc Networks. First International Conference, ADHOCNETS 2009, Niagara Falls, Ontario, Canada, September 22-25, 2009. Revised Selected Papers},
        proceedings_a={ADHOCNETS},
        year={2012},
        month={7},
        keywords={Security Intrusion detection Wormholes Wireless sensor networks Distributed systems},
        doi={10.1007/978-3-642-11723-7_14}
    }
    
  • Rennie Graaf
    Islam Hegazy
    Jeffrey Horton
    Reihaneh Safavi-Naini
    Year: 2012
    Distributed Detection of Wormhole Attacks in Wireless Sensor Networks
    ADHOCNETS
    Springer
    DOI: 10.1007/978-3-642-11723-7_14
Rennie Graaf1,*, Islam Hegazy1,*, Jeffrey Horton2,*, Reihaneh Safavi-Naini1,*
  • 1: University of Calgary
  • 2: University of Wollongong
*Contact email: rennie.degraaf@gmail.com, islam.hegazy@ucalgary.ca, jhorton@google.com, rei@ucalgary.ca

Abstract

Sensors in a wireless sensor network depend on their neighbours to route their messages. Yet, routing protocols in wireless sensor network are vulnerable to different types of attacks. In this paper, we consider the in which the adversary diverts traffic from one part of the network to another part by introducing a low cost tunnel between the two parts. We introduce a distributed intrusion detection system that monitors the communication in the network and propose a criterion for the placement of intrusion detection nodes. The intrusion detection system searches for violations of that criterion to detect wormholes of length above a certain minimum value. We evaluate the effectiveness of our system in a simulated environment. The experiments show that our system can detect 100% of the wormholes that are beyond the communication range of the intrusion detection nodes. Finally, we discuss our results and show directions for future work.