Research On Privacy Preserving Mechanism Of Spatial Crowdsourcing

The explosion of mobile communications and the Internet of Things(Io T)has spawned a new distributed computing paradigm,spatial crowdsourcing,in which workers actively participate in spatio-temporal based computing tasks to earn commissions,which facilitates the development of urban sharing economy services.Considering the mobile user side with limited storage space and computational power,spatial crowdsourcing platforms usually use the server assignment model(SAM)to directly assign tasks to workers,which means that crowdsourcing platforms extensively collect attribute information about tasks and workers,such as location and interests,to precisely implement task assignments.However,in the real world,crowdsourcing platforms are not completely trustworthy,and they may reveal sensitive information about workers and tasks,which will significantly cut down users’ interest in using spatial crowdsourcing services.Therefore,how to assign tasks efficiently and securely remains a serious problem to be solved.In spatial crowdsourcing,workers usually prefer to query for tasks that are nearest and satisfy their interests based on their locations.According to the above findings,the privacy-preserving spatial crowdsourcing mechanism based on location range query and multiple interest keyword query is concentrated.The main work is as follows.(1)A privacy-preserving scheme for task assignments(PPTA)based on the server assignment model is designed.First,the scheme utilizes inner product functional encryption to implement circular range queries and multi-keyword queries.Second,considering that workers usually tend to query the closest task to reduce travel costs,the scheme uses grid location intersection to sense the fine-grained location distance between tasks and workers.In addition,the scheme can implement user accountability and user revocation without updating the ciphertexts and keys,which enhances the practicality and security of the scheme to some extent.Finally,a server assignment model(SAM)algorithm is designed,which can effectively avoid the occurrence of assignment deadlock problems in multi-task and multi-worker scenarios.(2)A privacy-preserving scheme for task assignments based on edge computing(Pri TAEC)is designed,which combines obvious transfer(OT)and edge computing to achieve location privacy preservation in spatial crowdsourcing.First,the scheme implements rectangular range queries using Hilbert curves and Bloom filters.Then,the scheme uses Geohash location encoding and obvious transfer to achieve fine-grained location matching.Finally,an offline-online phase task assignment algorithm is designed to improve the efficiency of task assignments.