Real Time Outlier Detection in Wireless Sensor Networks
In the field of wireless sensor networks, the sensed patterns that significantly deviate from the normal patterns are considered as outliers. The possible sources of outliers include noise and errors, events, and malicious attacks on the network. Wireless Sensor Networks (WSNs) are more likely to generate outliers due to their special characteristics, like constrained available resources, frequent physical failure, and often harsh deployment area. In general, the outlier data are ignored, but in some critical applications like fire detection, hoax and intruder detection, these outlier data are much important to take decision in real time.