Blockchain and deep learning based trust management for Internet of Vehicles

【Author】 Wang, Shujuan; Hu, Yingnan; Qi, Guanqiu

【Source】SIMULATION MODELLING PRACTICE AND THEORY

【影响因子】4.199

【Abstract】Internet of Vehicles (IoVs) works as the most advanced component of Intelligent Transportation Systems (ITSs). In IoVs, vehicles are able to communicate with nearby vehicles or RoadSide Units (RSUs). Traffic safety and efficiency can be provided by collecting and uploading real-time traffic information through vehicles, as well as broadcasting information by RSUs. However, there may be malicious vehicles in the network uploading false information, which will lead to serious traffic problems. To alleviate this problem, a trust management system based on blockchain technology is proposed in this paper. In this system, vehicles in the network firstly collect information about their surroundings and then upload valid information to nearby RSUs. To prevent malicious vehicles from uploading false messages, this paper designs a deep learning based verification model to calculate the trustworthiness of uploaded messages, and to further obtain the credibility scores of vehicles using the calculated results, and detect malicious vehicles accordingly. Moreover, a public blockchain framework is proposed and a Proof-Of -Trust (POT) consensus algorithm is designed. Vehicles are motivated to report true and valid information, and are penalized for uploading false data under this framework. Simulation results show that this mechanism can effectively detect malicious vehicles and motivate unfamiliar vehicles to upload true and reliable information to achieve trust management in the open and dynamic vehicular network environments.

【Keywords】InternetofVehicles; Trustmanagement; Blockchain; Deeplearning

【发表时间】2022 NOV

【收录时间】2022-08-15

【文献类型】实验仿真

【主题类别】

区块链应用-实体经济-交通领域