Asymptotic Analysis of A Fuzzy Based Intrusion Detection System For Zigbee

Abstract

The prospects of achieving a trillion connected internet of things (IoT) devices by 2020 has created the urgency for effective intrusion detection systems (IDS) for these devices. Although it has been argued that the most effective technique used in such systems is anomaly detection, there exist no mechanisms to determine their performance in real-life deployment. In this paper, we report the results of applying asymptotic analysis to evaluate the performance of an anomaly detection algorithm which is designed using logic reasoning through fuzzy logic methodologies. In order to achieve this, the IDS was included as part of intrusion detection software for ZigBee Wireless Sensor Networks (WSNs). In particular, the solution is targeted to address the ZigBee protocol’s vulnerability to flood attacks during node discovery and association to the network. The intrusion detection software is hosted external to the WSNs in pursue of a light solution mindful of resource preservation in sensor nodes.