An Outlier Detection Scheme For Wireless Sensor Networks

Securing wireless sensor network is drawing a lot of attention of researchers with detecting outliers being a major concern. Outlier is referred to as deviated behavior from normal pattern of the network elements caused by noise, events or attacks. Outlier detection is crucial for normal functioning of the network and forms the first step towards security management. By taking suitable actions or by thwarting the erroneous behavior outliers can be managed. Events occur when there is a change in the state of the networks and is expected behavior in the network, whereas attacks are unexpected behavior in network causing difficult situation in the network. Event detection and attack detection are many a times mistook as same. Event detection is followed by suitable actions to handle the event whereas attack detection should be followed by counter measures to subdue the situation. In this paper, we propose an outlier detection scheme (ODS) for detecting events and attacks in the wireless sensor network. We evaluate the efficiency of scheme through simulations.