抵抗推理攻击的车联网位置隐私增强方法OA北大核心CSTPCD

Unclear entity credibility in the internet of vehicles(IoV),and frequent communication between users and other entities,can increase the risk of user privacy leakage.Therefore,a location privacy enhancement method against inference attacks in IoV is proposed.User can request the road side units(RSUs)for points of interest(POIs).The multiple suitable POIs are recommended based on user similarity from the local cache,so that the final destination among the multiple POIs is selected.Even if recommended POIs are stolen by an attacker,it is difficult to infer the user's final destination.If users do not find the desired POI,they can request the service provider(SP)to expand the service scope.The simulation results show that in comparison with the comparison method,the communication cost can decrease by 14%on average and the position entropy can increase by 12%on average.The proposed scheme can reduce communication costs and enhance privacy protection by reducing the number of communications with SP.Users can select the final destination by multiple POIs that are more consistent with users' expectations by means of user preference calculation,which can improve service availability and resist inference attacks.