@INPROCEEDINGS {qun_mobihoc07,
	author = {Bo Sheng and Qun Li and Weizhen Mao and Wen Jin},
	booktitle = {{ACM} International Symposium on Mobile Ad Hoc Networking and Computing ({Mobihoc})},
	PAGES = {219--228},
	TITLE = {Outlier Detection in Sensor Networks},
	Address ={Montreal, Canada},
	Month={Sept.},
	YEAR = {2007}
	}

Abstract 

Outlier detection has many important applications in sen-
sor networks, e.g., abnormal event detection, animal behav-
ior change, etc. It is a difficult problem since global infor-
mation about data distributions must be known to identify
outliers. In this paper, we use a histogram-based method
for outlier detection to reduce communication cost. Rather
than collecting all the data in one location for centralized
processing, we propose collecting hints (in the form of a his-
togram) about the data distribution, and using the hints to
filter out unnecessary data and identify potential outliers.
We show that this method can be used for detecting outliers
in terms of two different definitions. Our simulation results
show that the histogram method can dramatically reduce
the communication cost.