[Paper Review] Physical Layer Security of Autonomous Driving: Secure Vehicle-to-Vehicle Communication in A Security Cluster

Na Young Ahn, Dong Hoon Lee|arXiv (Cornell University)|Dec 13, 2019

Wireless Communication Security Techniques45 references10 citations

TL;DR

This paper proposes a physical layer security framework for vehicle-to-vehicle (V2V) communication in autonomous driving by introducing 'vehicular secrecy capacity'—a novel metric defined using SNR and vehicle-specific dynamics like speed, safety distance, and response time. The method enables real-time control of secrecy capacity in highway and urban environments, enhancing resilience against eavesdroppers, including quantum computers, through dynamic, physics-based security mechanisms.

ABSTRACT

We suggest secure Vehicle-to-Vehicle communications in a secure cluster. Here, the security cluster refers to a group of vehicles having a certain level or more of secrecy capacity. Usually, there are many difficulties in defining secrecy capacity, but we define vehicular secrecy capacity for the vehicle defined only by SNR values. Defined vehicular secrecy capacity is practical and efficient in achieving physical layer security in V2V. Typically, secrecy capacity may be changed by antenna related parameters, path related parameters, and noise related parameters. In addition to these conventional parameters, we address unique vehicle-related parameters, such as vehicle speed, safety distance, speed limit, response time, etc. in connection with autonomous driving. We confirm the relationship between vehicle-related secrecy parameters and secrecy capacity through modeling in highway and urban traffic situations. These vehicular secrecy parameters enable real-time control of vehicle secrecy capacity of V2V communications. We can use vehicular secrecy capacity to achieve secure vehicle communications from attackers such as quantum computers. Our research enables economic, effective and efficient physical layer security in autonomous driving.

Motivation & Objective

• To address the lack of practical physical layer security mechanisms in autonomous vehicle V2V communications.
• To define a new secrecy capacity metric tailored for vehicular networks using SNR and dynamic vehicle parameters.
• To enable real-time control of secrecy capacity by incorporating vehicle-specific dynamics such as speed, safety distance, and response time.
• To enhance security against advanced threats, including quantum computing attacks, through physical layer techniques.
• To provide an efficient, economic, and scalable solution for secure V2V communication in both highway and urban traffic scenarios.

Proposed method

• Proposes 'vehicular secrecy capacity' as a practical metric derived solely from SNR and vehicle dynamics, replacing conventional secrecy capacity definitions.
• Incorporates vehicle-specific parameters such as speed, safety distance, speed limit, and response time into the secrecy capacity model.
• Models secrecy capacity behavior in both highway and urban traffic environments using realistic propagation and mobility conditions.
• Uses signal-to-noise ratio (SNR) as the primary input for calculating secrecy capacity, enabling real-time adaptability.
• Applies physical layer security principles by leveraging channel state information and dynamic vehicle parameters to enhance confidentiality.
• Validates the framework through simulation-based modeling under diverse traffic scenarios to demonstrate dynamic secrecy capacity control.

Experimental results

Research questions

• RQ1 How can secrecy capacity in V2V communication be redefined to be practical and applicable in autonomous driving environments?
• RQ2 What role do vehicle-specific dynamics—such as speed, safety distance, and response time—play in shaping physical layer secrecy capacity?
• RQ3 How does the proposed vehicular secrecy capacity model perform under real-world highway and urban traffic conditions?
• RQ4 Can physical layer security be dynamically controlled in real time using only SNR and vehicle motion parameters?
• RQ5 To what extent can this framework resist advanced eavesdropping threats, including those from quantum computers?

Key findings

• The proposed vehicular secrecy capacity model successfully integrates SNR and vehicle dynamics into a single, practical metric for physical layer security.
• Vehicle speed, safety distance, and response time significantly influence the achievable secrecy capacity, enabling dynamic control.
• The model demonstrates stable and measurable secrecy capacity improvements in both highway and urban traffic scenarios.
• Real-time adaptability of secrecy capacity is achievable through continuous monitoring of SNR and vehicle motion parameters.
• The framework provides a foundation for resisting future threats such as quantum computing-based eavesdropping in V2V communications.
• The method is economically viable and scalable, offering an efficient alternative to conventional cryptographic methods in autonomous vehicle networks.