Enhancing CAN Security: A Fourier Transform Approach to Reverse Engineering

Modern vehicles are increasingly equipped with numerous electronic hardware devices, evolving into Connected and Autonomous Vehicles (CAVs). These devices, known as Electronic Control Units (ECUs), communicate via the Controller Area Network (CAN) protocol, a de-facto standard for inter-ECU communication. The relationships between CAN messages and vehicle functions are stored in CAN database (DBC) files, which are proprietary and typically withheld by Original Equipment Manufacturers (OEMs) for security reasons. Despite its reliability and cost-effectiveness, the CAN protocol’s broadcast nature makes it vulnerable to attackers who gain network access. Current defense solutions, such as Intrusion Detection Systems (IDS), are developed independently of DBC information, resulting in discrepancies between research threats and real-world vehicle threats. To address this, CAN message time difference calculation reversal methods can be employed, though time domain calculations often suffer from noise interference. This paper proposes a novel reverse engineering approach using Fourier Transform to perform calculations in the frequency domain, offering a more accurate analysis of CAN messages.