Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{199591,
        author = {D.V.RAJKUMAR and VISHNU .K and SATHISH M and VISHNUCHANDAR S.S},
        title = {Real Time Intrusion Detection System For Securing Automatic CAN Communication In Smart Vehicles},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {14412-14420},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=199591},
        abstract = {As modern vehicles transition into highly integrated cyber-physical systems, the security of internal communication architectures has become a paramount concern for automotive safety and reliability. The Controller Area Network (CAN) bus, which serves as the fundamental backbone for communication between Electronic Control Units (ECUs), was originally designed for real-time efficiency but lacks inherent security mechanisms such as message authentication and encryption.
This architectural vulnerability exposes vehicles to a broad spectrum of sophisticated cyberattacks, including Denial-of- Service (DoS) and malicious message injection, which can lead to the catastrophic failure of safety-critical functions. This research proposes an automated, real-time Intrusion Detection System (IDS) based on an unsupervised machine learning approach to mitigate these risks.
By utilizing the Isolation Forest algorithm, the proposed system establishes a statistical behavioral baseline by analyzing the inter- arrival times and frequency distribution of CAN frames. Unlike traditional rule-based security systems, this model is capable of identifying anomalies without requiring a labeled attack dataset, making it resilient against zero-day threats. Experimental results, conducted in a virtual CAN environment, demonstrate that the proposed hybrid detection framework achieves high sensitivity to frequency-based deviations with minimal computational latency.
The developed system provides a scalable and intelligent security layer that assists in proactive risk assessment while reducing the manual workload of cybersecurity professionals. This research significantly contributes to the early detection of automotive network intrusions, enhancing the overall resilience and integrity of modern vehicle electronic architectures.},
        keywords = {CAN Bus, Intrusion Detection System (IDS), Machine Learning, Isolation Forest, Automotive Cybersecurity, Anomaly Detection, Real-time Monitoring.},
        month = {April},
        }