Security Analysis and Improvement of Vehicle Ethernet SOME/IP Protocol

Area of Science:

• Automotive Cybersecurity
• Network Protocol Security
• Intelligent Transportation Systems

Background:

• Intelligent Connected Vehicles (ICVs) integrate in-vehicle networks and smart devices, enabling rapid communication.
• Increasingly complex external connections expose in-vehicle networks to significant security vulnerabilities.
• The security of in-vehicle communication protocols is critical for ICV safety and data integrity.

Purpose of the Study:

• To formally model and analyze the security of the SOME/IP protocol used in vehicle Ethernet.
• To identify and address specific vulnerabilities within the existing SOME/IP security protocol.
• To propose and validate an enhanced protocol with improved security mechanisms.

Main Methods:

• Modeling the SOME/IP protocol using Colored Petri Net (CPN) theory.
• Applying the Dolev-Yao adversary model for formal security verification and analysis.
• Implementing enhancements with timestamps and random numbers to mitigate identified threats.

Main Results:

• Formal verification revealed three key vulnerabilities in the SOME/IP protocol: replay, tampering, and deception attacks.
• The introduction of timestamps and random numbers effectively addresses these identified security weaknesses.
• The enhanced protocol demonstrates a significant improvement in overall security performance.

Conclusions:

• The CPN-based model and Dolev-Yao analysis provide a robust method for verifying in-vehicle network protocols.
• The proposed security enhancements for the SOME/IP protocol are effective against common cyber threats.
• Strengthening in-vehicle protocols is essential for ensuring the security and reliability of Intelligent Connected Vehicles.