International Journal of Advanced Research in Computer Science and Software Engineering (IJARCSSE)
In the field of wireless sensor networks, the measurements that deviate from the normal behavior of sensed data are taken to be as outliers. The potential sources of outliers can be noise and errors, events, and malicious attacks on the network. This paper gives an overview of existing outlier detection techniques specifically developed for the wireless sensor networks. Also, a technique-based taxonomy will be discussed with characteristics of outlier data such as data type, outlier type, outlier identity, and outlier degree.