A Mahalanobis-based data fusion and anomaly filtering method for ensuring information integrity in autonomous vehicle populations

J.A. Pavelina, I.Y. Popov

Abstract


This paper examines the problem of ensuring information integrity in V2X networks during information exchange between groups of autonomous vehicles (AVs). This problem arises in a dynamic environment and the presence of saboteurs implementing False Data Injection (FDI) attacks, in which classical methods of linear averaging of sensor data lead to critical control errors. A data fusion method with adaptive weighting and outlier filtering based on the Mahalanobis distance is proposed. A comparative experiment of the proposed method with approaches based on Euclidean distance and one-dimensional Z-score is conducted. The results demonstrate that the proposed method maintains high state estimation accuracy and trajectory stability even when up to 20% of the group's nodes are compromised.

Full Text:

PDF (Russian)

References


Arena F., Pau G. An overview of vehicular communications //Future internet. – 2019. – Т. 11. – №. 2. – С. 27.

Pavelina J.A., Mukhamedzhanov S., I.I. Komarov I.I. Method for Building an Environment Model of Autonomous Vehicles // International Journal of Open Information Technologies. – 2024. – Т. 12. – № 7. – С. 38-42.

Lu Z., Qu G., Liu Z. A survey on recent advances in vehicular network security, trust, and privacy //IEEE Transactions on Intelligent Transportation Systems. – 2018. – Т. 20. – №. 2. – С. 760-776.

ALASEM R., Mansour M. Decentralized Trust Model for Vehicle Ad-Hoc Networks (VANETs) with 5G Integration: A Blockchain-Based Approach for Enhanced Security and Privacy in Intelligent Transportation Systems. – 2025.

Aggarwal C. C., Sathe S. Which outlier detection algorithm should I use? //Outlier Ensembles: An Introduction. – Cham : Springer International Publishing, 2017. – С. 207-274.

Rousseeuw P. J., Hubert M. Anomaly detection by robust statistics //Wiley interdisciplinary reviews: Data mining and knowledge discovery. – 2018. – Т. 8. – №. 2. – С. e1236.

Wang F. et al. A multivariate time series anomaly detection model based on spatio-temporal dual features //2023 International Conference on Networking and Network Applications (NaNA). – IEEE, 2023. – С. 416-421.

Pavelina J.A. A method for determining the trustworthiness of autonomous vehicles based on social forces // International Journal of Open Information Technologies. – 2026. – Т. 14. – № 2. – С. 141-146.


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность Monetec 2026 СНЭ

ISSN: 2307-8162