The School of Computer Science is pleased to present…
Intrusion Detection of sophisticated safety-related threats in Controller Area Network (CAN) bus using Machine Learning
MSc Thesis Proposal by: Vikrant
Date: Friday, 12 July 2024
Time: 1:00 pm
Location: Essex Hall, room 122
Abstract: Electronic Control Units (ECUs) have become important to modern vehicles, enhancing operational control, driving comfort, and safety. These ECUs communicate using the Controller Area Network (CAN) protocol, which, despite its widespread adoption, is vulnerable to various security threats. According to the Upstream 2023 report, the number of high and massive-scale incidents that potentially impacted thousands of mobility assets increased by x2.5 compared to 2022. While developing a more secure CAN protocol is crucial, intrusion detection systems (IDS) offer a viable path to mitigating cyberattacks on vehicles. This research proposes a machine learning-based IDS employing a range of models, including K-Nearest Neighbor (KNN), Logistic Regression, Decision Tree (DT), Random Forest, XGBoost, CatBoost, Multi-Layer Perceptron (MLP), Long Short-Term Memory (LSTM), and Transformer-based approaches. Our research used the newly published can-train-test dataset specifically designed for building ML-based IDS. The dataset consists of nine different types of attacks performed on four different vehicles in a real-world environment. This study majorly focused on DoS, Fuzzy, Gear Spoofing, Speed Spoofing and Standstill attacks performed on Chevrolet Impala and Subaru Forester. Notably, our IDS, which leverages both traditional machine learning models and advanced deep learning architectures, achieves good accuracy with less training and prediction time, exhibiting a high true positive rate and a low false negative rate. Additionally, our novel approach involves converting CAN bus data into feature strings for classification using transformers-based models, providing enhanced interpretability and robustness. The comparative evaluation of our results indicates that our proposed IDS surpasses existing methods in terms of reliability and efficiency, effectively detecting cyber-attacks with minimal error rates.
Keywords: Vehicle Security, Controller Area Network, Intrusion Detection, Machine Learning, Transformers Models
Thesis Committee:
Internal Reader: Dr. Muhammad Assaduzzaman
External Reader: Dr. Mohammad Hassanzadeh
Advisor: Dr. Ikjot Saini