Title :
Passive human presence detection method based on the incremental analysis of principal components from the signal strength space
Author :
Nenadic, D. ; Mrazovac, Bojan
Author_Institution :
Fakultet Tehnickih Nauka, Novi Sad, Serbia
Abstract :
This paper proposes the use of the incremental method for principal components analysis, designed for human presence detection. The algorithm extracts principal components from the matrix of signal strength samples retrieved from the pre-existing wireless electrical infrastructure, based on smart outlets and smart light switches. By proposing the additional filter based on the Shannon Entropy, the level of false detections is minimized.
Keywords :
entropy; filtering theory; principal component analysis; signal detection; Shannon entropy; false detection; filter; incremental analysis; passive human presence detection method; principal component analysis; signal strength sample; signal strength space; smart light switch; smart outlet; wireless electrical infrastructure; Covariance matrix; Electronic mail; Humans; IEEE 802.15 Standards; Principal component analysis; Wireless sensor networks; Zigbee; CCIPCA; PCA; RSSI; Zigbee; pasivna detekcija prisustva čoveka;
Conference_Titel :
Telecommunications Forum (TELFOR), 2012 20th
Conference_Location :
Belgrade
Print_ISBN :
978-1-4673-2983-5
DOI :
10.1109/TELFOR.2012.6419242
