Selfishness detection for backoff algorithms in wireless networks

  • Authors:
  • Antoniy Ganchev;Lata Narayanan

  • Affiliations:
  • Department of Computer Science and Software Engineering, Concordia University, Montreal, Canada;Department of Computer Science and Software Engineering, Concordia University, Montreal, Canada

  • Venue:
  • WIMOB '11 Proceedings of the 2011 IEEE 7th International Conference on Wireless and Mobile Computing, Networking and Communications
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Selfish nodes in an 802.11 network can gain unfair access to the wireless medium by modifying the backoff protocol, for example by choosing smaller backoff values more often than would be dictated by pure chance. Detecting this kind of misbehavior is far from obvious as it is not always possible to deduce the backoff values used by a node. We propose a new backoff scheme called XVBEB in which there are only two backoff values: 0 and CW. We describe how to deduce the backoff values used by an observed node using XVBEB based on observations of transmissions by nodes in the network and the collision timeline. Given a set of backoff values used by a XVBEB node, we describe how to conclude with a specified level of certainty whether the node is indeed adhering to the protocol. We also show that it would take much more effort to detect selfishness for 802.11 nodes following the standard backoff procedure within a comparable misbehaving framework.