Design and Analysis of Resilient Vehicular Platoon Systems over Wireless Networks

TL;DR

This paper presents a resilient framework for vehicular platoons over wireless networks, focusing on clock synchronization and attack mitigation to enhance safety and reliability in traffic management.

Contribution

It introduces a diffusion-based re-synchronization protocol and a new resilience metric, advancing the robustness of vehicular platoons against clock attacks and transmission delays.

Findings

Resilient design maintains synchronization despite increased delay variance.

Reliability improves by 45% over baseline methods.

Feasibility demonstrated with nine-fold delay variance increase.

Abstract

Connected vehicular platoons provide a promising solution to improve traffic efficiency and ensure road safety. Vehicles in a platoon utilize on-board sensors and wireless vehicle-to-vehicle (V2V) links to share traffic information for cooperative adaptive cruise control. To process real-time control and alert information, there is a need to ensure clock synchronization among the platoon's vehicles. However, adversaries can jeopardize the operation of the platoon by attacking the local clocks of vehicles, leading to clock offsets with the platoon's reference clock. In this paper, a novel framework is proposed for analyzing the resilience of vehicular platoons that are connected using V2V links. In particular, a resilient design based on a diffusion protocol is proposed to re-synchronize the attacked vehicle through wireless V2V links thereby mitigating the impact of variance of the transmission delay during recovery. Then, a novel metric named temporal conditional mean exceedance is defined and analyzed in order to characterize the resilience of the platoon. Subsequently, the conditions pertaining to the V2V links and recovery time needed for a resilient design are derived. Numerical results show that the proposed resilient design is feasible in face of a nine-fold increase in the variance of transmission delay compared to a baseline designed for reliability. Moreover, the proposed approach improves the reliability, defined as the probability of meeting a desired clock offset error requirement, by 45% compared to the baseline.