Traffic Signal Intrusion Detection

Our research primarily focused on developing sophisticated Intrusion Detection Systems (IDS) to safeguard traffic signals and Intelligent Transportation Systems (ITS) from various cyber threats. Moving beyond the conventional techniques that rely on data from connected vehicles and image analysis, our objective was to create a robust multi-layered detection approach considering a wider set of traffic parameters and anomalies.

Project Details:

  • An Evidence Theoretic Approach for Traffic Signal Intrusion Detection: Our aim was to extend the realm of intrusion detection by integrating anomalies in flow rate, phase time, and vehicle speed. Using Dempster–Shafer decision theory and Shannon’s entropy, we measured the uncertainty of observations. We validated this approach using a traffic simulator called SUMO, real scenarios, and historical data, achieving a detection accuracy of 79.3% and a reduced rate of false alarms.
  • Information Fusion-based Cybersecurity Threat Detection for Intelligent Transportation System: We addressed the dynamic and complex nature of ITS and irregular traffic patterns by creating an information fusion-based detection method. Our system efficiently classified traffic conditions by integrating the Kalman filter for noise reduction, Dempster-Shafer decision theory, and Shannon’s entropy. Our simulation results showcased that the fusion of data from three sensors was particularly effective in detecting normal traffic conditions, while the fusion of two sensor data yielded better results for anomaly detection.
  • Detecting Intrusion in the Traffic Signals of an Intelligent Traffic System: An integral part of our project was developing an intrusion detection technique that utilized the flow rate and phase time of a traffic signal to detect intrusions. We processed this information using the Dempster-Shafer (DS) theory. Our proposed IDS was tested in a range of traffic situations, including intrusion scenarios, with results showing a high detection accuracy of over 91%.

Project Outcome:

The outcome of these research projects was a comprehensive and highly effective Intrusion Detection System capable of securing traffic signals and Intelligent Transportation Systems. The system demonstrated robustness in dealing with different types of cyberattacks and showed high detection accuracy while minimizing false alarms. It proved highly adaptable in dealing with the dynamic and complex nature of traffic patterns, contributing to safer and more reliable transportation networks. Our findings and methodologies are detailed in the following papers:

  • Chowdhury, Abdullahi, Gour Karmakar, Joarder Kamruzzaman, Rajkumar Das, and SH Shah Newaz. “An Evidence Theoretic Approach for Traffic Signal Intrusion Detection.” Sensors 23, no. 10 (2023): 4646.
  • Chowdhury, Abdullahi, Gour Karmakar, Joarder Kamruzzaman, and Tapash Saha. “Detecting intrusion in the traffic signals of an intelligent traffic system.” In Information and Communications Security: 20th International Conference, ICICS 2018, Lille, France, October 29-31, 2018, Proceedings, pp. 696-707. Springer International Publishing, 2018.
  • Chowdhury, Abdullahi, Ranesh Naha, Shahriar Kaisar, Ali Khoshkholghi, Kamran Ali, and Antonino Galletta. “Information fusion-based cybersecurity threat detection for the intelligent transportation system.” In The 23rd International Symposium on Cluster, Cloud and Internet Computing (CCGrid 2023), Bangalore, India | May 1-4, 2023.

The collective efforts of our team, including Dr. Abdullahi Chowdhury, Assoc. Prof. Gour Karmakar, Prof. Joarder Kamruzzaman, Dr. Rajkumar Das, Dr. SH Shah Newaz, Tapash Saha, Dr. Ranesh Naha, Dr. Shahriar Kaisar, Dr. Ali Khoshkholghi, Dr. Kamran Ali, and Prof. Antonino Galletta, significantly contributed to the success of this project.

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *