Using A One-Class Compound Classifier To Detect In-Vehicle Network Attacks

Andrew John Tomlinson, Jeremy Bryans, Siraj Shaikh

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

    17 Citations (Scopus)
    120 Downloads (Pure)

    Abstract

    The Controller Area Network (CAN) in vehicles provides serial communication between electronic control units that manage en- gine, transmission, steering and braking. Researchers have recently demonstrated the vulnerability of the network to cyber-attacks which can manipulate the operation of the vehicle and compromise its safety. Some proposals for CAN intrusion detection systems, that identify attacks by detecting packet anomalies, have drawn on one-class classi cation, whereby the system builds a decision surface based on a large number of normal instances. The one-class approach is discussed in this paper, together with initial results and observations from implementing a classi er new to this eld. The Compound Classier has been used in image processing and medical analysis, and holds advantages that could be relevant to CAN intrusion detection.
    Original languageEnglish
    Title of host publicationProceedings of the Genetic and Evolutionary Computation Conference Companion
    PublisherAssociation for Computing Machinery (ACM)
    Pages1926-1929
    Number of pages4
    ISBN (Print)978-1-4503-5764-7
    DOIs
    Publication statusPublished - 6 Jul 2018
    EventThe Genetic and Evolutionary Computation Conference - Kyoto, Japan
    Duration: 15 Jul 201819 Jul 2018
    http://gecco-2018.sigevo.org/index.html/tiki-index.php

    Conference

    ConferenceThe Genetic and Evolutionary Computation Conference
    Abbreviated titleGECCO 2018
    Country/TerritoryJapan
    CityKyoto
    Period15/07/1819/07/18
    Internet address

    Bibliographical note

    © ACM, 2018. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in Proceedings of the Genetic and Evolutionary Computation Conference Companion
    http://doi.acm.org/10.1145/3205651.3208223

    Keywords

    • intrusion detection
    • nearest neighbour
    • classifier
    • cybersecurity
    • anomaly detection
    • one-class
    • controller area network

    Fingerprint

    Dive into the research topics of 'Using A One-Class Compound Classifier To Detect In-Vehicle Network Attacks'. Together they form a unique fingerprint.

    Cite this