(Semi-)Automatically Parsing Private Protocols for In-Vehicle ECU Communications

Area of Science:

• Automotive Cybersecurity
• Embedded Systems Engineering
• Network Protocol Analysis

Background:

• In-vehicle electronic control unit (ECU) communications rely on proprietary protocols over Controller Area Network (CAN) specifications.
• Understanding these private protocols is crucial for assessing vehicle security and developing effective Intrusion Detection and Prevention Systems (IDPS).

Purpose of the Study:

• To propose and validate a suite of methods for parsing proprietary ECU protocols within in-vehicle CAN networks.
• To enhance the security analysis and IDPS design for modern vehicles.

Main Methods:

• Development of algorithms for parsing discrete and continuous variables transmitted over CAN.
• Implementation of a parsing method leveraging upper-layer protocols like On-Board Diagnostics (OBD) and Unified Diagnostic Services (UDS).
• Extensive verification across five diverse vehicle brands, including an electric vehicle.

Main Results:

• Demonstrated universality and correctness of the proposed parsing algorithms across multiple vehicle brands.
• Achieved semi-automatic parsing for continuous variables and fully automatic parsing for upper-layer protocols.
• The parsing results serve as a key indicator for penetration testing on in-vehicle CAN networks.

Conclusions:

• The developed methods provide a significant advancement in analyzing and securing in-vehicle communication protocols.
• The approach offers a practical solution for cybersecurity researchers and automotive engineers.
• The findings contribute to building more robust and secure automotive systems against cyber threats.