Sybil Attack Resilient Traffic Networks: A Physics-Based Trust Propagation Approach

Abstract

We study a crowdsourcing aided road traffic estimation setup, where a fraction of users (vehicles) are malicious, and report wrong sensory information, or even worse, report the presence of Sybil (ghost) vehicles that do not physically exist. The motivation for such attacks lies in the possibility of creating a "virtual" congestion that can influence routing algorithms, leading to "actual" congestion and chaos. We propose a Sybil attack-resilient traffic estimation and routing algorithm that is resilient against such attacks. In particular, our algorithm leverages noisy information from legacy sensing infrastructure, along with the dynamics and proximity graph of vehicles inferred from crowdsourced data. Furthermore, the scalability of our algorithm is based on efficient Boolean Satisfiability (SAT) solvers. We validated our algorithm using real traffic data from the Italian city of Bologna. Our algorithm led to a significant reduction in average travel time in the presence of Sybil attacks, including cases where the travel time was reduced from about an hour to a few minutes.