Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system

<p>Modern automobiles are equipped with a large number of electronic control units (ECUs) to provide safe, driver assistance and comfortable services. The controller area network (CAN) provides near real-time data transmission between ECUs with adequate reliability for in-vehicle communication...

Full description

Saved in:
Bibliographic Details
Main Author: Sampath Rajapaksha (17541411) (author)
Other Authors: Harsha Kalutarage (17541414) (author), M. Omar Al-Kadri (17541417) (author), Andrei Petrovski (6262691) (author), Garikayi Madzudzo (17541420) (author)
Published: 2023
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1864513535736807424
author Sampath Rajapaksha (17541411)
author2 Harsha Kalutarage (17541414)
M. Omar Al-Kadri (17541417)
Andrei Petrovski (6262691)
Garikayi Madzudzo (17541420)
author2_role author
author
author
author
author_facet Sampath Rajapaksha (17541411)
Harsha Kalutarage (17541414)
M. Omar Al-Kadri (17541417)
Andrei Petrovski (6262691)
Garikayi Madzudzo (17541420)
author_role author
dc.creator.none.fl_str_mv Sampath Rajapaksha (17541411)
Harsha Kalutarage (17541414)
M. Omar Al-Kadri (17541417)
Andrei Petrovski (6262691)
Garikayi Madzudzo (17541420)
dc.date.none.fl_str_mv 2023-09-01T09:00:00Z
dc.identifier.none.fl_str_mv 10.1016/j.jisa.2023.103570
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Beyond_vanilla_Improved_autoencoder-based_ensemble_in-vehicle_intrusion_detection_system/24717189
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Engineering
Automotive engineering
Information and computing sciences
Artificial intelligence
Cybersecurity and privacy
Machine learning
Controller Area Network (CAN)
Machine learning
Automotive cybersecurity
Deep learning
Autoencoder
Anomaly detection
dc.title.none.fl_str_mv Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p>Modern automobiles are equipped with a large number of electronic control units (ECUs) to provide safe, driver assistance and comfortable services. The controller area network (CAN) provides near real-time data transmission between ECUs with adequate reliability for in-vehicle communication. However, the lack of security measures such as authentication and encryption makes the CAN bus vulnerable to cyberattacks, which affect the safety of passengers and the surrounding environment. Detecting attacks on the CAN bus, particularly masquerade attacks, presents significant challenges. It necessitates an intrusion detection system (IDS) that effectively utilizes both CAN ID and payload data to ensure thorough detection and protection against a wide range of attacks, all while operating within the constraints of limited computing resources. This paper introduces an ensemble IDS that combines a gated recurrent unit (GRU) network and a novel autoencoder (AE) model to identify cyberattacks on the CAN bus. AEs are expected to produce higher reconstruction errors for anomalous inputs, making them suitable for anomaly detection. However, vanilla AE models often suffer from overgeneralization, reconstructing anomalies without significant errors, resulting in many false negatives. To address this issue, this paper proposes a novel AE called Latent AE, which incorporates a shallow AE into the latent space. The Latent AE model utilizes Cramér’s statistic-based feature selection technique and a transformed CAN payload data structure to enhance its efficiency. The proposed ensemble IDS enhances attack detection capabilities by leveraging the best capabilities of independent GRU and Latent AE models, while mitigating the weaknesses associated with each individual model. The evaluation of the IDS on two public datasets, encompassing 13 different attacks, including sophisticated masquerade attacks, demonstrates its superiority over baseline models with near real-time detection latency of 25ms.</p><h2>Other Information</h2> <p> Published in: Journal of Information Security and Applications<br> License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.jisa.2023.103570" target="_blank">https://dx.doi.org/10.1016/j.jisa.2023.103570</a></p>
eu_rights_str_mv openAccess
id Manara2_9bf1fb2d6da56d9e4fe632185b546351
identifier_str_mv 10.1016/j.jisa.2023.103570
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/24717189
publishDate 2023
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection systemSampath Rajapaksha (17541411)Harsha Kalutarage (17541414)M. Omar Al-Kadri (17541417)Andrei Petrovski (6262691)Garikayi Madzudzo (17541420)EngineeringAutomotive engineeringInformation and computing sciencesArtificial intelligenceCybersecurity and privacyMachine learningController Area Network (CAN)Machine learningAutomotive cybersecurityDeep learningAutoencoderAnomaly detection<p>Modern automobiles are equipped with a large number of electronic control units (ECUs) to provide safe, driver assistance and comfortable services. The controller area network (CAN) provides near real-time data transmission between ECUs with adequate reliability for in-vehicle communication. However, the lack of security measures such as authentication and encryption makes the CAN bus vulnerable to cyberattacks, which affect the safety of passengers and the surrounding environment. Detecting attacks on the CAN bus, particularly masquerade attacks, presents significant challenges. It necessitates an intrusion detection system (IDS) that effectively utilizes both CAN ID and payload data to ensure thorough detection and protection against a wide range of attacks, all while operating within the constraints of limited computing resources. This paper introduces an ensemble IDS that combines a gated recurrent unit (GRU) network and a novel autoencoder (AE) model to identify cyberattacks on the CAN bus. AEs are expected to produce higher reconstruction errors for anomalous inputs, making them suitable for anomaly detection. However, vanilla AE models often suffer from overgeneralization, reconstructing anomalies without significant errors, resulting in many false negatives. To address this issue, this paper proposes a novel AE called Latent AE, which incorporates a shallow AE into the latent space. The Latent AE model utilizes Cramér’s statistic-based feature selection technique and a transformed CAN payload data structure to enhance its efficiency. The proposed ensemble IDS enhances attack detection capabilities by leveraging the best capabilities of independent GRU and Latent AE models, while mitigating the weaknesses associated with each individual model. The evaluation of the IDS on two public datasets, encompassing 13 different attacks, including sophisticated masquerade attacks, demonstrates its superiority over baseline models with near real-time detection latency of 25ms.</p><h2>Other Information</h2> <p> Published in: Journal of Information Security and Applications<br> License: <a href="http://creativecommons.org/licenses/by/4.0/" target="_blank">http://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1016/j.jisa.2023.103570" target="_blank">https://dx.doi.org/10.1016/j.jisa.2023.103570</a></p>2023-09-01T09:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1016/j.jisa.2023.103570https://figshare.com/articles/journal_contribution/Beyond_vanilla_Improved_autoencoder-based_ensemble_in-vehicle_intrusion_detection_system/24717189CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/247171892023-09-01T09:00:00Z
spellingShingle Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
Sampath Rajapaksha (17541411)
Engineering
Automotive engineering
Information and computing sciences
Artificial intelligence
Cybersecurity and privacy
Machine learning
Controller Area Network (CAN)
Machine learning
Automotive cybersecurity
Deep learning
Autoencoder
Anomaly detection
status_str publishedVersion
title Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
title_full Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
title_fullStr Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
title_full_unstemmed Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
title_short Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
title_sort Beyond vanilla: Improved autoencoder-based ensemble in-vehicle intrusion detection system
topic Engineering
Automotive engineering
Information and computing sciences
Artificial intelligence
Cybersecurity and privacy
Machine learning
Controller Area Network (CAN)
Machine learning
Automotive cybersecurity
Deep learning
Autoencoder
Anomaly detection