Intrusion Detection System for CAN Bus In-Vehicle Network based on Machine Learning Algorithms

Abstract

The advent of automotive industry has an increasing market demand pertaining to installation of intelligent transportation facilities in modern vehicles. Now more comfortable and safer travelling experience is one best trait of these vehicles. Moreover, it has opened new gates to advancement in automotive sector. Modern vehicles are connected to advance systems and technologies using various communication protocols. Amongst numerous communication protocols, one widely used protocol is the controller area network (CAN) bus which serves as a central medium for in-vehicle communications. However, the communication in these vehicles may impose greater threats and may ultimately compromise the security by breaching the system. Various attacks on CAN bus may compromise the confidentiality, integrity and availability of vehicular data through intrusions which may endanger the physical safety of vehicle and passengers. In this paper, a novel machine learning based approach is used to devise an Intrusion Detection System for the CAN bus network. The proposed system is scalable and adaptable to a diverse set of emerging attacks on autonomous vehicles. Results witnessed the accuracy of 100% of our proposed system in detecting and safeguarding threats against multiple impersonation and denial of service attacks as well as 99% accuracy of fuzzy attacks.