Behavioural Intrusion Detection for Wireless Sensor Networks

Wireless sensor networks are used increasingly in various different fields. The information which they measure and distribute is often sensitive or valuable and this incentivises attacks on the networks. Moreover, they are distributed and make use of wireless communications which makes them vulnerable to attack. This paper presents an intrusion detector for wireless sensor networks which makes use of behavioural data from the network. The intrusion detectors was evaluated in a resource constrained environment because wireless sensor networks have limited computational and energy resources. The developed system was found to be very accurate as well as computationally lightweight.