Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

65 Citations (Scopus)
119 Downloads (Pure)

Abstract

Outlier detection in wireless sensor networks is essential to ensure data quality, secure monitoring and reliable detection of interesting and critical events. A key challenge for outlier detection in wireless sensor networks is to adaptively identify outliers in an online manner with a high accuracy while maintaining the resource consumption of the network to a minimum. In this paper, we propose one-class support vector machine-based outlier detection techniques that sequentially update the model representing normal behavior of the sensed data and take advantage of spatial and temporal correlations that exist between sensor data to cooperatively identify outliers. Experiments with both synthetic and real data show that our online outlier detection techniques achieve high detection accuracy and low false alarm rate.
Original languageUndefined
Title of host publicationProceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia
Place of PublicationBradford, United Kingdom
PublisherIEEE Computer Society Press
Pages990-995
Number of pages6
ISBN (Print)978-0-7695-3639-2
DOIs
Publication statusPublished - 26 May 2009
Event23rd IEEE International Conference on Advanced Information Networking and Applications, AINA 2009 - Bradford, United Kingdom
Duration: 26 May 200929 May 2009
Conference number: 23
http://www.inf.brad.ac.uk/~iawan/aina/

Publication series

Name
PublisherIEEE Computer Society Press

Workshop

Workshop23rd IEEE International Conference on Advanced Information Networking and Applications, AINA 2009
Abbreviated titleAINA
CountryUnited Kingdom
CityBradford
Period26/05/0929/05/09
Internet address

Keywords

  • METIS-263863
  • IR-65500
  • EWI-15391

Cite this

Zhang, Y., Meratnia, N., & Havinga, P. J. M. (2009). Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks. In Proceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia (pp. 990-995). Bradford, United Kingdom: IEEE Computer Society Press. https://doi.org/10.1109/WAINA.2009.200
Zhang, Y. ; Meratnia, Nirvana ; Havinga, Paul J.M. / Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks. Proceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia. Bradford, United Kingdom : IEEE Computer Society Press, 2009. pp. 990-995
@inproceedings{18c9f7e598f24ba988680aade363ee2c,
title = "Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks",
abstract = "Outlier detection in wireless sensor networks is essential to ensure data quality, secure monitoring and reliable detection of interesting and critical events. A key challenge for outlier detection in wireless sensor networks is to adaptively identify outliers in an online manner with a high accuracy while maintaining the resource consumption of the network to a minimum. In this paper, we propose one-class support vector machine-based outlier detection techniques that sequentially update the model representing normal behavior of the sensed data and take advantage of spatial and temporal correlations that exist between sensor data to cooperatively identify outliers. Experiments with both synthetic and real data show that our online outlier detection techniques achieve high detection accuracy and low false alarm rate.",
keywords = "METIS-263863, IR-65500, EWI-15391",
author = "Y. Zhang and Nirvana Meratnia and Havinga, {Paul J.M.}",
year = "2009",
month = "5",
day = "26",
doi = "10.1109/WAINA.2009.200",
language = "Undefined",
isbn = "978-0-7695-3639-2",
publisher = "IEEE Computer Society Press",
pages = "990--995",
booktitle = "Proceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia",

}

Zhang, Y, Meratnia, N & Havinga, PJM 2009, Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks. in Proceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia. IEEE Computer Society Press, Bradford, United Kingdom, pp. 990-995, 23rd IEEE International Conference on Advanced Information Networking and Applications, AINA 2009, Bradford, United Kingdom, 26/05/09. https://doi.org/10.1109/WAINA.2009.200

Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks. / Zhang, Y.; Meratnia, Nirvana; Havinga, Paul J.M.

Proceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia. Bradford, United Kingdom : IEEE Computer Society Press, 2009. p. 990-995.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks

AU - Zhang, Y.

AU - Meratnia, Nirvana

AU - Havinga, Paul J.M.

PY - 2009/5/26

Y1 - 2009/5/26

N2 - Outlier detection in wireless sensor networks is essential to ensure data quality, secure monitoring and reliable detection of interesting and critical events. A key challenge for outlier detection in wireless sensor networks is to adaptively identify outliers in an online manner with a high accuracy while maintaining the resource consumption of the network to a minimum. In this paper, we propose one-class support vector machine-based outlier detection techniques that sequentially update the model representing normal behavior of the sensed data and take advantage of spatial and temporal correlations that exist between sensor data to cooperatively identify outliers. Experiments with both synthetic and real data show that our online outlier detection techniques achieve high detection accuracy and low false alarm rate.

AB - Outlier detection in wireless sensor networks is essential to ensure data quality, secure monitoring and reliable detection of interesting and critical events. A key challenge for outlier detection in wireless sensor networks is to adaptively identify outliers in an online manner with a high accuracy while maintaining the resource consumption of the network to a minimum. In this paper, we propose one-class support vector machine-based outlier detection techniques that sequentially update the model representing normal behavior of the sensed data and take advantage of spatial and temporal correlations that exist between sensor data to cooperatively identify outliers. Experiments with both synthetic and real data show that our online outlier detection techniques achieve high detection accuracy and low false alarm rate.

KW - METIS-263863

KW - IR-65500

KW - EWI-15391

U2 - 10.1109/WAINA.2009.200

DO - 10.1109/WAINA.2009.200

M3 - Conference contribution

SN - 978-0-7695-3639-2

SP - 990

EP - 995

BT - Proceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia

PB - IEEE Computer Society Press

CY - Bradford, United Kingdom

ER -

Zhang Y, Meratnia N, Havinga PJM. Adaptive and Online One-Class Support Vector Machine-based Outlier Detection Techniques for Wireless Sensor Networks. In Proceedings of the IEEE 23rd International Conference on Advanced Information Networking and Applications Workshops/Symposia. Bradford, United Kingdom: IEEE Computer Society Press. 2009. p. 990-995 https://doi.org/10.1109/WAINA.2009.200