Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks

Unsupervised anomaly detection (UAD) techniques have received increasing attention in wireless sensor networks (WSNs). However, the high dimensional training data often make sensor nodes unable to sustain in computation, and result in quite expensive communication overhead. The feature reduction tec...

Full description

Bibliographic Details
Main Authors: Xie, Miao, Han, Song, Tian, Biming
Other Authors: Yi Mu
Format: Conference Paper
Published: IEEE 2011
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/25234
_version_ 1848751652007313408
author Xie, Miao
Han, Song
Tian, Biming
author2 Yi Mu
author_facet Yi Mu
Xie, Miao
Han, Song
Tian, Biming
author_sort Xie, Miao
building Curtin Institutional Repository
collection Online Access
description Unsupervised anomaly detection (UAD) techniques have received increasing attention in wireless sensor networks (WSNs). However, the high dimensional training data often make sensor nodes unable to sustain in computation, and result in quite expensive communication overhead. The feature reduction techniques make great sense through the reduction of the dimensionality when the features are strongly interrelated. Among these UAD techniques, distance-based anomaly detection (DB-AD) is a special one that allows to be described by a probability model. Based on this observation, DB-AD is explored deeply with a feature reduction technique, principal component analysis (PCA). Through examining the proportion of the variance explained by the first principal component (PC), a new feature reduction approach is proposed for DB- AD in WSNs, which enables to reduce the dimensionality to one in any situation. Specifically, the first PC is alone used for representing the original data as long as it retains most of the variance; otherwise, the information loss is geometrically reverted to neutralize the error. By obtaining a tradeoff between the detection error and performance overload, this approach is significant for resource-constrained WSNs, as the computational complexity and communication overhead will be reduced to a fraction of the original magnitude. Finally, this approach is evaluated with a real WSN dataset.
first_indexed 2025-11-14T07:56:07Z
format Conference Paper
id curtin-20.500.11937-25234
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T07:56:07Z
publishDate 2011
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-252342023-01-27T05:26:33Z Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks Xie, Miao Han, Song Tian, Biming Yi Mu Gregorio Martinez Geyong Min anomaly detection feature reduction wireless sensor networks Unsupervised anomaly detection (UAD) techniques have received increasing attention in wireless sensor networks (WSNs). However, the high dimensional training data often make sensor nodes unable to sustain in computation, and result in quite expensive communication overhead. The feature reduction techniques make great sense through the reduction of the dimensionality when the features are strongly interrelated. Among these UAD techniques, distance-based anomaly detection (DB-AD) is a special one that allows to be described by a probability model. Based on this observation, DB-AD is explored deeply with a feature reduction technique, principal component analysis (PCA). Through examining the proportion of the variance explained by the first principal component (PC), a new feature reduction approach is proposed for DB- AD in WSNs, which enables to reduce the dimensionality to one in any situation. Specifically, the first PC is alone used for representing the original data as long as it retains most of the variance; otherwise, the information loss is geometrically reverted to neutralize the error. By obtaining a tradeoff between the detection error and performance overload, this approach is significant for resource-constrained WSNs, as the computational complexity and communication overhead will be reduced to a fraction of the original magnitude. Finally, this approach is evaluated with a real WSN dataset. 2011 Conference Paper http://hdl.handle.net/20.500.11937/25234 10.1109/TrustCom.2011.73 IEEE restricted
spellingShingle anomaly detection
feature reduction
wireless sensor networks
Xie, Miao
Han, Song
Tian, Biming
Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks
title Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks
title_full Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks
title_fullStr Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks
title_full_unstemmed Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks
title_short Highly efficient distance-based anomaly detection through univariate with PCA in wireless sensor networks
title_sort highly efficient distance-based anomaly detection through univariate with pca in wireless sensor networks
topic anomaly detection
feature reduction
wireless sensor networks
url http://hdl.handle.net/20.500.11937/25234