Outlier Detection in Wireless Sensor Networks Based on Copula theory

Abstract

Wireless sensor network has progressively emerged as a leading technology to develop new wireless applications that can revolutionize and improve our daily lives. Nonetheless, many WSN applications are mission-critical such as military and disaster area applications. Also, the quality of the sensed data stream collected by sensor nodes is affected by anomalies that occur due to various reasons like noise and errors, events, and malicious attacks. Thereby, anomaly detection in sensor readings efficiently is a major concern and imperative for making right decisions. In this paper, we address these challenges and we propose a new fully on-line distributed approach for anomaly detection in WSN based on Copula theory. Our experimental results on real datasets collected by real sensor network prove the efficiency and effectiveness of the proposed approach and, show that the proposed approach can achieve high detection accuracy with a low false alarm rate