Research On Conditional Privacy-preserving Authentication And Protocol Automatic Analysis Model In VANETs

Under the guidance and promotion of the 13 th Five-Year Plan,the intellectualization and networking of automobiles will become the inevitable trend of the development in the future.Vehicular ad hoc networks(VANETs)have great development prospects as the foundation of intelligent transportation.Good deployment of VANETs can provide support for intelligent traffic management and effectively promote the construction of intelligent cities.Although VANETs has a widely application prospect,the openness of communication channel let message interaction extremely vulnerable to being eavesdropped and modified in VANETs,Which threatens the security of the VANETs system.A large number of identity authentication and message authentication protocols have been proposed and used to solve this problem.However,traditional authentication protocol will lead to the disclosure of the privacy data of vehicle and users.Although these data seem to be unrelated and can be shared openly,with the development of big data and artificial intelligence technology in recent years,attackers can analyze the privacy information such as the daily life habits of vehicle users in the case of obtaining large amounts of data.Therefore,under the premise of ensuring VANETs message authentication,how to guarantee user data privacy is a difficult problem to be solved urgently.Dealing with the above problems,this paper has carried out in-depth research on the conditional privacy-preserving authentication for the vehicle ad hoc,and designed a conditional anonymous authentication scheme based on mobile infrastructure.The proposed using urban public traffic vehicles instead of traditional roadside unit(RSU)to authenticates vehicle,As buses have the characteristics of wide coverage,fixed routes,on-demand dynamic optimization,etc.it can effectively reduce constant switching between vehicles and roadside units.Secondly,the scheme solves the problem of vehicle effectively revocation.The system master key is stored in the tamper-proof device of the bus.The legitimate user vehicle can obtain the pseudo identity and corresponding private key when broadcasting messages through interaction with the bus,and the revocation user cannot successfully apply.Moreover,the process does not require the participation of a trusted center(TA),thus effectively relieving the computational pressure of TA.At the same time,aiming at the situation that the batch message contains invalid signatures,the scheme proposes a binary sorting tree method to extract valid signatures from the batch message,thus improving the efficiency of batch authentication.Thirdly,the security of the proposed is proved by the formalization method.Finally,in order to demonstrate the performance and feasibility of the proposed conditional anonymous authentication scheme,we have analyzed the proposed performance.And further simulates through ns2 and sumo joint simulation platform.The results show that the proposed has good performance in terms of the end-to-end delay and packet loss ratio.On the other hand,When proving the security of the protocol,we find that the traditional security method(such as BAN logic)has high requirements for the prior knowledge of protocol analyst,and it is difficult to analyze the security of large-scale protocols rapidly.After studying,we find that the computing power of the vehicle nodes is relatively weak,the VANETs conditional anonymous authentication protocol itself is relatively simple,the protocol dataset is huge,and the attack type can be classified.In view of the above features,we proposes a novel XGBoost-based automatic security analysis method for VANETs authentication protocol,Through feature extraction and vectorization of the collected vehicle network authentication protocol and the authentication protocol in this paper,we construct three diffirent types of protocol dataset models,and use XGBoost to train datasets,so that to achieve the purpose of correct classification.In the end,we evaluate the automatic analysis method by experiments on the indicators such as accuracy and recall ratio,and the results show that the proposed method is feasible and effective.