A blockchain-based reputation system for trusted VANET nodes

Vehicular network applications, such as Local Danger Warnings (LDW), have become necessary as they provide services to drivers with warnings and instructions to ensure safe driving. However, before any vehicle reacts to an alert, it is essential to verify its veracity because an incorrect warning misleads other vehicles and causes unnecessary delays and detours in traffic. On the other hand, ignoring an alert or taking too long to verify it can cause accidents. A common way to address this problem is via a reputation system, where the veracity and plausibility of an alert message are assessed based on the reputation of the vehicle that discloses it. This paper describes a decentralized reputation system based on a consortium blockchain and smart contracts called BRS4VANETs. This system analyzes the reliability of data generated by a vehicle, detects malicious behavior, and contributes to decision-making. The complete functional architecture of the system, detailing the algorithms, evaluation metrics, communication protocol, and message formats, is described and implemented. Experiments with network and traffic simulators evaluated the risk of a false message attack, the effectiveness of the system and its overheads. The assessment demonstrates the importance of reputation systems and also shows the feasibility of having a decentralized system to store and disseminate reputation information based on blockchain technologies. Finally, the results also demonstrate that BRS4VANETs can successfully detect suspicious and malicious vehicles.