| With the continuous development of Internet of Vehicles and communication technology,location-based services and various services derived from Internet of Vehicles are widely used,such as query of interest points,navigation services,Mobile Crowdsensing services of Internet of Vehicles and so on.These services bring people endless convenience in daily life and make earth-shaking changes in society.However,when enjoying these services,users’ privacy is facing many threats.According to the location information of users,service providers will constantly analyze private content,which will even threaten the safety of users’ lives and property.Therefore,ensuring the location privacy of Internet of Vehicles users is of great significance to the healthy development of Internet of Vehicles industry.Geo-indistinguishability has become an important standard for location privacy protection in recent years.Unfortunately,the existing location privacy protection schemes based on geo-indistinguishability in the Internet of Vehicles environment often ignore the accuracy of query,resulting in the increase of users’ travel distance,and then affect the enthusiasm of workers to participate in services.In addition,in the continuous query mode,the linear combination mechanism of geo-indistinguishability itself cannot effectively protect the location privacy of users.In view of the above problems,this paper introduces Johnson-Lindenstrauss,caching,edge computing and other technologies,combined with the geo-indistinguishability mechanism,and designs the location privacy protection mechanism in the scenario of the Internet of Vehicles Mobile Crowdsensing and continuous query.The main research work of this paper is as follows:In the Internet of Vehicles Mobile Crowdsensing environment,the geoindistinguishability mechanism often has a great impact on the accuracy of task allocation,resulting in the increase of users’ travel distance.This paper proposes a safe and efficient task allocation model in the Internet of Vehicles Mobile Crowdsensing.The model uses Johnson Lindenstrauss algorithm to transform the location of users and task points in the same dimension,so as to realize location privacy protection and accurate task allocation.Moreover,the model adopts the architecture of edge computing to reduce the overhead of communication by aggregating requests in groups.Finally,in terms of security,the model meets the differential privacy of location data set.In terms of the accuracy of task allocation,compared with the traditional scheme of differential privacy and geo-indistinguishability,while ensuring privacy and security,this scheme can achieve more than 90% of the accuracy of task allocation.In the continuous query scenario of the Internet of Vehicles,the linear combination mechanism of geo-indistinguishability will lead to the rapid consumption of users’ privacy cost,which will not provide effective security.This paper proposes a geoindistinguishability scheme based on cooperative cache,which reduces the number of communications with the server by using the cache capacity of the vehicle,so as to reduce the privacy consumption caused by using geo-indistinguishability,and further improves the hit rate of cache through the strategy of multi-level propagation.In order to speed up the efficiency of cache update,the minimum heap storage method is used to store the cache,and the least recently used strategy is used to update the cache,which greatly reduces the time overhead caused by cache update.Finally,the experimental results show that compared with the traditional scheme of geo-indistinguishability and the scheme of geo-indistinguishability based on prediction mechanism,this scheme has more advantages in saving privacy overhead,saving about 15% privacy overhead,and can respond to users’ query requests faster. |
