id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_xjfxdj37xfdyjcezj2nj3cnaki	Daniel-Ioan Curiac	Ensemble based sensing anomaly detection in wireless sensor networks	2012	11	.pdf	application/pdf	8281	799	59	Ensemble based sensing anomaly detection in wireless sensor networks This paper tackles the sensing anomaly detection from a new perspective by modeling the correct operation of sensors not by one, but by five different ensemble based system composed of a set of diverse binary classifiers, adequately selected, to implement Moreover, every time a sensing anomaly is discovered, our ensemble offers a reliable estimation to replace the erroneous measurement provided by From this perspective we developed five classifiers that combine the two kinds of input data (measurement time series provided by the sensor under Classifier's autoregressive prediction block receives past measurement values provided by the sensor under investigation: In case a measurement provided by the sensor under investigation is considered abnormal, the ensemble decision block offers a (data not used in ensemble training) obtained through computer simulations or using experimental sensor deployments in a real environment.	./cache/work_xjfxdj37xfdyjcezj2nj3cnaki.pdf	./txt/work_xjfxdj37xfdyjcezj2nj3cnaki.txt