Periodic Guarding for Energy-Efficient Anomaly Detection in Wireless Sensor Networks

摘要

Anomaly detection is an important challenge for applications such as monitoring, fault diagnosis and intrusion detection wireless sensor networks (WSNs) are envisaged to support. The algorithms developed for anomaly detection have to consider the inherent limitations of WSNs in their design so that the communication overhead in sensor nodes is minimized and the lifetime of the network is maximized. To address this problem, we propose a novel, distributed rule-based anomaly detection algorithm, which exploits the peculiar characteristics of the IEEE 802.15.4 standard in order to incarnate the concept of energy-efficient periodic guarding for anomaly detection purposes. Simulation based studies would demonstrate that the proposed algorithm achieves very high detection accuracy, while keeping the communication overhead at very low levels.