| With the development of the mobile Internet and the popularity of mobile smart devices in recent years,the Internet of Things(Io T)has flourished.As an emerging application in the Io T,mobile crowdsensing(MCS)has attracted much attentions on academia and industry,and has been widely used in environmental monitoring,traffic monitoring and other scenarios due to its high efficiency,convenience and low energy consumption.In MCS,task requesters publish tasks to the sensing platform,and the sensing platform assigns these tasks to workers with mobile devices.In general,the sensing platform matches tasks and workers according to the location and time attributes of tasks and workers.However,most current studies rarely consider location and time attributes,simultaneously.In addition,in some involving geographical location,task assignment in MCS may involve the location privacy of participating users,and users may be threatened by the leakage of these sensitive information.Therefore,it is necessary to achieve real-time task assignment while protecting the location privacy of participating users in MCS.We adopt and online tasking strategy to achieve the real-time goal,the sensing platform sequentially matches the arrived tasks or workers.Geo-indistinguishability is used to protect the privacy of the geographic location,which can protect the privacy of user's locations against adversaries with arbitrary background knowledge.However,it will cause serious errors if the sensing platform directly uses the locations generated by geo-indistinguishability to assign tasks.Therefore,it is of great significance to explore how to assign tasks in real time while protecting user location privacy using geo-indistinguishability.In view of the above problems,this thesis has done the following work for task assignment in MCS:1.This thesis proposes two real-time task assignment schemes with location privacy protection.The first scheme(Scheme-?)only protects the location privacy of worker,and the second scheme(Scheme-?)is to protect the location privacy of workers and tasks,simultaneously,and real-time task allocation is realized in these two schemes.Both schemes use geo-indistinguishability to protect location privacy,and achieve privacy protection without relying on trusted third parties and against background knowledge attacks.2.In the scheme for protecting the location privacy of workers,we first quantify the reachability between the worker with obfuscated location and the task with actual location through the probabilistic method.Then,in order to reduce the worker's travel distance and the communication overhead of the scheme,we design corresponding distance comparison mechanisms for workers and tasks arriving at the sensing platform,namely the worker-based distance comparison mechanism(WDCM)and the task-based distance comparison mechanism(TDCM).WDCM is used to select the task which is closest to the arrived worker in probability,and TDCM is used to select the worker who is probably closest to the arrived task.3.Scheme-? for protecting the privacy of workers and tasks is extension of Scheme-? for protecting the privacy of workers.Both of them protect the privacy of workers,while Scheme-? protects the location privacy of sensitive tasks.In Scheme-?,we first quantify the reachability between the worker and the task that both use obfuscated location,and then design the workerbased distance comparison mechanism(WDCM-?)and the task-based distance comparison mechanism(TDCM)to reduce the travel distance and reduce system overhead.4.In order to verify the effectiveness of two real-time task assignment schemes with location privacy protection proposed in this thesis,we conduct a series of comparative experiments on the two schemes based on real datasets and analyze the experimental results in detail.The experimental results show the feasibility and effectiveness of Scheme-? and Scheme-?,and have lower computational overhead and communication overhead. |
PDF Full Text Request