earticle

영어

This paper proposes a distributed, lightweight and efficient method for the detection of attacks against data traffic flow. It was designed for wireless sensor networks (WSNs), which contain a large number of static sensor nodes that periodically transmit the sensed data. The intrusion detection scheme is based on analysis of the incoming traffic and a fuzzy inference system. During the first stage, a lightweight thresholding technique is applied to the selected traffic attributes – the packet reception rate (PRR) and packet inter-arrival time (IAT) – in a time window of a given length. Then, the fuzzy inference system derives its final conclusion on the basis of both criteria, deciding whether an attack is taking place. The proposed method does not require installation of any additional hardware or extra communication costs. Its computational requirements are negligible, since it is based on lightweight statistical methods. The proposed scheme may be applied to both hierarchical and flat WSNs. The proposed technique was evaluated in Castalia simulator for WSNs, and demonstrated better accuracy in comparison with common algorithms. It may serve as a component of a complex intrusion detection system (IDS).