Abstract
Wireless sensor networks (WSNs) are important platforms for collecting environmental data and monitoring phenomena. Anomalies caused by hardware and software errors, unusual events, and malicious attacks affect the integrity of data gathered by such networks. Therefore, anomaly detection process is a necessary step in building sensor network systems to assure data quality for right decision making. This paper presents a new distributed online anomaly detection model that measures the dissimilarity of sensor observations in principal component space. The new model distributes the detection process over the network to minimize energy consumption while ensuring high detection effectiveness. The detection effectiveness and efficiency of the new model are proved through experiments on real world dataset from Sensorscope system project. Experimental results reveal that the model achieved high detection rate with relatively few false positive rates compared to an existing model.