A Group-Vehicles Oriented Reputation Assessment Scheme for Edge VANETs

With the development of the smart traffic, the traditional vehicular Ad hoc Networks (VANETs) and Traffic Estimation and Prediction System (TrEPS) do not satisfy the growing safety requirement, due to the network delay, transmit price and privacy security. In this paper, we propose a group-vehicles oriented reputation assessment scheme for edge VANETs. Firstly, based on edge computing, we build a reputation assessment framework for Group-Vehicles, to validate the correctness of message for other vehicles rapidly. Secondly, through filtering the malicious feedback and faulty message, our scheme can effectively defend against the Bad-mouth attack and Zigzag attack to assure the security of VANETs. Thirdly, the message isolation is implemented by the group-vehicles management, to enhance the privacy security of scheme. In the end, we validate the effectiveness of our scheme through experiments. In other words, even though the proportion of Bad-mouth attack vehicles is about 40%, the precision is 92.12%, and the recall is 88.25%. Also, the proportion of Zigzag attack vehicles is about 40%, the precision is 88.52%, and the recall is 86.75%.