TY - GEN
T1 - DAdAE
T2 - 49th Annual Conference of the IEEE Industrial Electronics Society, IECON 2023
AU - Mavikumbure, Harindra S.
AU - Cobilean, Victor
AU - Wickramasinghe, Chathurika S.
AU - Varghese, Benny J.
AU - Pennington, Timothy
AU - Manic, Milos
N1 - Funding Information:
ACKNOWLEDGEMENTS The Department of Energy partly supported this work through the U.S. DOE Idaho Operations Office under Contract DE-AC07-05ID14517, and partly by the Commonwealth Cyber Initiative, an Investment in the Advancement of Cyber Research and Development, Innovation and Workforce Development (cyberinitiative.org).
Publisher Copyright:
© 2023 IEEE.
PY - 2023/11/16
Y1 - 2023/11/16
N2 - Modern vehicles have multiple electronic control units (ECUs) that are connected as part of a complex cyber-physical system (CPS). The controller area network (CAN) is a well-known communication protocol that connects these ECUs because of its reliability and efficiency. However, adversaries can easily inject abnormal messages into the CAN bus remotely to affect vehicle driving safety. Existing anomaly detection methods only focus on specific vehicle models and have a limited range of applications across different vehicles. To address this challenge, this paper proposes a Domain Adversarial training-based AutoEncoder (DAdAE) for unsupervised CAN anomaly detection. The advantages of our approach are: 1) detect variant attack scenarios on different car models 2) does not require labeled data 3) works well even with a limited dataset. The effectiveness of the proposed model is evaluated on the survival dataset, and the experiment results show that the DAdAE model improves the overall f1 score significantly, compared to other unsupervised models.
AB - Modern vehicles have multiple electronic control units (ECUs) that are connected as part of a complex cyber-physical system (CPS). The controller area network (CAN) is a well-known communication protocol that connects these ECUs because of its reliability and efficiency. However, adversaries can easily inject abnormal messages into the CAN bus remotely to affect vehicle driving safety. Existing anomaly detection methods only focus on specific vehicle models and have a limited range of applications across different vehicles. To address this challenge, this paper proposes a Domain Adversarial training-based AutoEncoder (DAdAE) for unsupervised CAN anomaly detection. The advantages of our approach are: 1) detect variant attack scenarios on different car models 2) does not require labeled data 3) works well even with a limited dataset. The effectiveness of the proposed model is evaluated on the survival dataset, and the experiment results show that the DAdAE model improves the overall f1 score significantly, compared to other unsupervised models.
KW - Anomaly Detection
KW - CAN bus communication
KW - Deep Learning
KW - Domain Adaptation
KW - Domain Adversarial Learning
KW - In-Vehicle Communications
UR - https://www.scopus.com/pages/publications/85179512811
UR - https://www.mendeley.com/catalogue/979de91e-8ba2-30c5-b886-bf4f74950be4/
U2 - 10.1109/IECON51785.2023.10312001
DO - 10.1109/IECON51785.2023.10312001
M3 - Conference contribution
AN - SCOPUS:85179512811
SN - 9798350331820
T3 - IECON Proceedings (Industrial Electronics Conference)
SP - 1
EP - 7
BT - IECON 2023 - 49th Annual Conference of the IEEE Industrial Electronics Society
PB - IEEE Computer Society
Y2 - 16 October 2023 through 19 October 2023
ER -