Distributed online outlier detection in wireless sensor networks using ellipsoidal support vector machine

    Research output: Contribution to journalArticleAcademicpeer-review

    58 Citations (Scopus)

    Abstract

    Low quality sensor data limits WSN capabilities for providing reliable real-time situation awareness. Outlier detection is a solution to ensure the quality of sensor data. An effective and efficient outlier detection technique for WSNs not only identifies outliers in a distributed and online manner with high detection accuracy and low false alarm, but also satisfies WSN constraints in terms of communication, computational and memory complexity. In this paper, we take into account the correlation between sensor data attributes and propose two distributed and online outlier detection techniques based on a hyperellipsoidal one-class support vector machine (SVM). We also take advantage of the theory of spatio-temporal correlation to identify outliers and update the ellipsoidal SVM-based model representing the changed normal behavior of sensor data for further outlier identification. Simulation results show that our adaptive ellipsoidal SVM-based outlier detection technique achieves better detection accuracy and lower false alarm as compared to existing SVM-based techniques designed for WSNs.
    Original languageUndefined
    Pages (from-to)1062-1074
    Number of pages13
    JournalAd hoc networks
    Volume11
    Issue number3
    DOIs
    Publication statusPublished - May 2013

    Keywords

    • EWI-23088
    • METIS-296308
    • Ellipsoidal support vector machine
    • Wireless Sensor Networks
    • IR-84331
    • Spatial correlation
    • Temporal correlation
    • METIS-296769
    • Outlier Detection

    Cite this

    @article{06869da1a1c24fa7bcedf49efdf04d3f,
    title = "Distributed online outlier detection in wireless sensor networks using ellipsoidal support vector machine",
    abstract = "Low quality sensor data limits WSN capabilities for providing reliable real-time situation awareness. Outlier detection is a solution to ensure the quality of sensor data. An effective and efficient outlier detection technique for WSNs not only identifies outliers in a distributed and online manner with high detection accuracy and low false alarm, but also satisfies WSN constraints in terms of communication, computational and memory complexity. In this paper, we take into account the correlation between sensor data attributes and propose two distributed and online outlier detection techniques based on a hyperellipsoidal one-class support vector machine (SVM). We also take advantage of the theory of spatio-temporal correlation to identify outliers and update the ellipsoidal SVM-based model representing the changed normal behavior of sensor data for further outlier identification. Simulation results show that our adaptive ellipsoidal SVM-based outlier detection technique achieves better detection accuracy and lower false alarm as compared to existing SVM-based techniques designed for WSNs.",
    keywords = "EWI-23088, METIS-296308, Ellipsoidal support vector machine, Wireless Sensor Networks, IR-84331, Spatial correlation, Temporal correlation, METIS-296769, Outlier Detection",
    author = "Y. Zhang and Nirvana Meratnia and Havinga, {Paul J.M.}",
    year = "2013",
    month = "5",
    doi = "10.1016/j.adhoc.2012.11.001",
    language = "Undefined",
    volume = "11",
    pages = "1062--1074",
    journal = "Ad hoc networks",
    issn = "1570-8705",
    publisher = "Elsevier",
    number = "3",

    }

    Distributed online outlier detection in wireless sensor networks using ellipsoidal support vector machine. / Zhang, Y.; Meratnia, Nirvana; Havinga, Paul J.M.

    In: Ad hoc networks, Vol. 11, No. 3, 05.2013, p. 1062-1074.

    Research output: Contribution to journalArticleAcademicpeer-review

    TY - JOUR

    T1 - Distributed online outlier detection in wireless sensor networks using ellipsoidal support vector machine

    AU - Zhang, Y.

    AU - Meratnia, Nirvana

    AU - Havinga, Paul J.M.

    PY - 2013/5

    Y1 - 2013/5

    N2 - Low quality sensor data limits WSN capabilities for providing reliable real-time situation awareness. Outlier detection is a solution to ensure the quality of sensor data. An effective and efficient outlier detection technique for WSNs not only identifies outliers in a distributed and online manner with high detection accuracy and low false alarm, but also satisfies WSN constraints in terms of communication, computational and memory complexity. In this paper, we take into account the correlation between sensor data attributes and propose two distributed and online outlier detection techniques based on a hyperellipsoidal one-class support vector machine (SVM). We also take advantage of the theory of spatio-temporal correlation to identify outliers and update the ellipsoidal SVM-based model representing the changed normal behavior of sensor data for further outlier identification. Simulation results show that our adaptive ellipsoidal SVM-based outlier detection technique achieves better detection accuracy and lower false alarm as compared to existing SVM-based techniques designed for WSNs.

    AB - Low quality sensor data limits WSN capabilities for providing reliable real-time situation awareness. Outlier detection is a solution to ensure the quality of sensor data. An effective and efficient outlier detection technique for WSNs not only identifies outliers in a distributed and online manner with high detection accuracy and low false alarm, but also satisfies WSN constraints in terms of communication, computational and memory complexity. In this paper, we take into account the correlation between sensor data attributes and propose two distributed and online outlier detection techniques based on a hyperellipsoidal one-class support vector machine (SVM). We also take advantage of the theory of spatio-temporal correlation to identify outliers and update the ellipsoidal SVM-based model representing the changed normal behavior of sensor data for further outlier identification. Simulation results show that our adaptive ellipsoidal SVM-based outlier detection technique achieves better detection accuracy and lower false alarm as compared to existing SVM-based techniques designed for WSNs.

    KW - EWI-23088

    KW - METIS-296308

    KW - Ellipsoidal support vector machine

    KW - Wireless Sensor Networks

    KW - IR-84331

    KW - Spatial correlation

    KW - Temporal correlation

    KW - METIS-296769

    KW - Outlier Detection

    U2 - 10.1016/j.adhoc.2012.11.001

    DO - 10.1016/j.adhoc.2012.11.001

    M3 - Article

    VL - 11

    SP - 1062

    EP - 1074

    JO - Ad hoc networks

    JF - Ad hoc networks

    SN - 1570-8705

    IS - 3

    ER -