Summary
This paper investigates the application of artificial intelligence and machine learning techniques to detect intrusions in Internet of Vehicles (IoV) communications, specifically targeting vulnerabilities in the widely-used Controller Area Network (CAN) protocol. Using a realistic dataset of vehicular traffic under benign and malicious conditions, the authors systematically evaluate 25 machine learning models and identify ensemble-based and tree-based approaches as superior performers for handling imbalanced data. The work provides a methodological framework for testing detection models and offers practical recommendations for improving the robustness of security solutions in real-world vehicular network environments.
Regional applicability
The findings are applicable to UK vehicular cybersecurity policy and practice, though they address a general technical challenge (network intrusion detection) rather than UK-specific regulatory or operational contexts. Implementation would depend on UK adoption of these AI-driven detection approaches in automotive and connected vehicle standards.
Key measures
Accuracy, balanced accuracy, F1-score, computational efficiency
Outcomes reported
The study evaluated 25 machine learning models for detecting cyberattacks in Controller Area Network (CAN) communications using the CICIoV2024 dataset, measuring performance across accuracy, balanced accuracy, F1-score, and computational efficiency metrics.
Topic tags
Dig deeper with Pulse AI.
Pulse AI has read the whole catalogue. Ask about this record, its theme, or how the findings apply to UK farming and policy — every answer cites the underlying studies.