Research On Privacy Protection Issues In Internet Of Vehicles

Through the analysis and processing of Safety Beacon Message(SBM),the Internet of Vehicles(Io V)can effectively improve the high safety traffic and intelligent traffic management.However,with the development and progress of social civilization,users pay more and more attention to personal privacy information.Therefore,in the process of SBM data acquisition,transmission and processing,privacy protection of user identity,location and trajectory is a hot issue that needs to be solved urgently in the current Io V applications.With the rapid iterative update of information technology,the application scope of Io V has been continuously expanded.New data-centric applications,such as 5G mobile communication network,crowdsourcing and crowdsensing,are emerging in Io V.However,due to the limitation of too many third-party entities,the traditional centralized architecture of Io V has the risk of privacy disclosure.It hinders the further development of these data-centric applications.Moreover,these existing privacy protection schemes are inadaptable to the feature requirements of these data-centric applications,such as the dynamism of 5G,the orientation of crowdsourcing,and the relevance of crowdsensing in Io V.And they can not effectively protect the privacy of user’s identity,location and trajectory.Therefore,this thesis is devoted to studying how to combine blockchain and edge computing technology to build a distributed architecture of Io V.The proposed architecture solves the privacy leakage caused by the insecure storage and transmission of SBM data.We design the identity and location privacy protection algorithm on the user layer of the proposed architecture.It solves the problem of privacy leakage caused by data transparency after SBM data is stored in blockchain.Then,according to these characteristics of data-centric applications,such as dynamism,orientation and relevance,we designe adaptive privacy protection schemes respectively to ensure privacy protection in the process of SBM data acquisition and transmission.The main research contents of this thesis are as follows:(1)This thesis has proposed an edge-assistant network architecture based on blockchain,including user layer,edge layer and central control layer.In the proposed architecture,we dempoly the enhanced edge node(Edge-Node)with semantic analysis and learning ability.It realizes the rapid retrieval and classification processing of distributed and massive SBM data.By defining three different data structures(the microblock,the critical block,and the validation block),we design the master-slave multi-chain model based on hash anchor.It gets through the SBM data of the whole network,and ensures the consistency and integrity of data by blockchain.Thus,it has solved the privacy leakage caused by insecure data storage.Between users and Edge-Nodes,we design the data upload and data access mechanisms based on attribute encryption.They have solved the privacy leakage caused by insecure data transmission outside the blockchain.(2)In the user layer of the edge-assistant network architecture based on blockchain,we design the identity and location privacy protection algorithms,which are used to strengthen the privacy protection before SBM data is on blockchain.It solves the problem of user privacy leakage caused by blockchain data transparency.First,we design LPP algorithm and obtain vehicle anonymity group.By defining the average connectivity(35)and average distance D,the effectiveness of k-anonymity unity is quantified in vehicle anonymity group.It improves the anonymity of vehicle location and strengthens the location privacy protection of users.On the basis of vehicle anonymity group,we then design the IPP algorithm.According to the dynamic threshold encryption method,the IPP algorithm divides the identity into multiple sub-identities.By updating and combing the sub-identities,it greatly improves the privacy protection of user’s identity.(3)In view of the dynamic characteristic of Io V 5G application,we have proposed the privacy metric composite index KDT,and designed the dynamic privacy protection algorithm DGD based on hot spot.First,to enable the vehicles having dynamic connectivity in 5G,the vehicle group model based on Mobile Femtocell(MFemtocell)is establised.Then,the DGD algorithm divides the locations into individual and social hotspots.By combing the location characteristic and MFemtocell technology,it dynamically generates the vehicle group.The vehicles in social hotspots have priority to automatically access the vehicle group.In this way,the vehicle group has more trajectory crossings and higher trajectory complexity.It can protect users’trajectory privacy more effectively.In the vehicle group,by defining the tracking probability p_iand pseudonym entropy H,we improve effectiveness of vehicle pseudonym exchange.Thus,the DGD algorithm cuts off the connection between vehicle identity and location/trajectory,enhancing the protection of location and trajectory privacy in dynamic environment.(4)In view of the orientation characteristic of Io V crowdsourcing application,we design time-tolerance anonymous privacy protection TAA algorithm and K-1location-offset privacy protection KLO algorithm.We establish two models of SBM data transmission(data aggregation and data distribution),and analyze location attack model of directional vehicle under the condition of identity information disclosure.TAA algorithm is designed in the process of data aggregation.In a time-tolerant way,TTA uploads SBM data with group identity.It cuts off the relationship between vehicle and SBM data acquisition location l_o,thus protecting the location privacy of directional vehicle in the process of data aggregatiion.According to the maximum offset probability entropy principle,KLO algorithm blurs the request location l_r into k POI positions.It proteces the location privacy of directional vehicle in the process of data distribution.(5)In view of the relevance characteristic of Io V crowdsensing application,we design a privacy protection CAPP algorithm based on correlation analysis.Considering the large amount,redundancy and correlation of crowdsensing data,the vehicle group is constructed by combining anonymity and homogeneity.According to the relationship between continuous locations and continuous vehicle groups,the identity management strategy is designed to realize the identity privacy protection of de-correlation.By analyzing the temporal,spatial and data correlation between the continuous locations,a privacy protection model based on CRF is constructed to solve the conditional probability distribution of locations.Then,we design the location suppression function in CAPP,and the vehicle can suppress SBM data uploading at some special locations with a certain probability.It realizesε-privacy protection of location while ensuring zero information loss.