The Research Of Privacy-preserving Task Allocation For Spatial Crowdsourcing

Spatial crowdsourcing(SC)is a popular type of crowdsourcing.In spatial crowdsourcing,all SC tasks are associated with locations/regions,and given an SC task,only the participants in the specified location/region are allowed to submit their answers of the task.However,outsourcing the SC tasks to a remote SC server poses new security concerns.Traditional SC server requires task requesters and participants to upload their locations to calculate the distances and complete the task allocation.As personal privacy,the location should not be exposed to any untrustworthy entities(even the SC server).Therefore,privacy protection is important in SC.How to balance the trade-off between efficiency and privacy has become an urgent problem to be solved.In addition,the previous solutions cannot achieve personalized customization of the task region,which makes the SC task inflexible and unable to meet the needs of task requesters.Aiming at the privacy preservation of the location and personal information of task requesters and participants in the task allocation of SC,this thesis proposes a novel SC privacy preservation framework,Beta-SC.The main contents of this thesis are as follows:(1)This thesis focus on the adaptive region representation accuracy control of SC task region.First of all,this thesis employs the prefix membership verification to represent the SC task regions and participants' locations to a set of prefixes.It evaluates if a location is inside an SC task region by evaluating if a prefix-based Boolean expression is true or not.Therefore,the adaptive region representation accuracy control is achieved in Beta-SC.(2)This thesis expresses the prefix-based Boolean expression as an access policy tree,and associate it with a policy-hidden attribute-based encryption scheme,so the task region privacy is preserved.In addition,this thesis utilizes a hybrid encryption scheme to achieve the privacy preservation of the task region and the location of the participants and the privacy preservation of the SC task contents and answers at the same time.Theoretical analysis and extensive experiments both show that Beta-SC framework has higher security than the previous schemes,and a high computational efficiency.(3)This thesis employs the emerging Blockchain technologies to handle a malicious SC server.The design of the Blockchain can provide the trusted and irreversible task publishing record,by deploying smart contracts on the Ethereum Blockchain to defend or detect the attacks from malicious SC server.And this thesis is verified by the experiment of the operational cost for using Ethereum Blockchain.The experimental results show that publishing SC task has an acceptable cost in Beta-SC framework,and it can be applied to real-world SC applications.