Systematic Attack Analysis and Adaptive Security in V2V Networks
Primary Investigator:
Bharat Bhargava
Bharat Bhargava, Pelin Angin, Miguel Villarreal-Vasquez, Amber Johnson, Gisele Munyengabe, Denis Ulybyshev
Abstract
Vehicle-to-vehicle (V2V) communication allows vehicles to communicate, enabling interactions with other vehicles and with the roadside infrastructure, which provides access to the backbone network. Vehicles exchange data related to traffic congestions, and safety warnings, which have been turned into hot targets of attackers, and thus are getting much attention to researchers. In this work, we systematically synthesize different types of V2V attacks, e.g., GPS spoofing and hidden vehicle attacks, and identify the attack features, the mitigation techniques and their costs, the impact on security and also the impact on safety. However, mitigating all such attacks and ensuring security and privacy affect vehicle safety because security and privacy measures include additional latency to the transmission of all types of V2V messages. Since different types of V2V messages have different levels of sensitivity to security and privacy and different security measures of different configuration parameters for a secure channel result in different performance overhead, we propose an adaptive security model for V2V networks that changes the configuration parameters of the secure channel dynamically based on the sensitivity of the V2V message, safety level of vehicles, and also the current network context. We use open source traffic simulation tools SUMO and TraNS to implement our solution and conduct analysis and experiments.