Research On Privacy And Security In Vehicle Crowd-sensing

Involved from traditional sensor network and data crowdsource,mobile crowdsensing as a new data collecting paradigm plays a key role in Internet of Things(IoT)and Internet of Vehicle(IoV),because most IoV applications depend on amount of vehicle sensing data.And various sensor devices and smart data processing applications enable vehicles to implement more complex sensing tasks.Therefore,it is worthy researching on vehicle-oriented crowdsensing.Design of vehicle crowdsensing framework should take the ability of system response and privacy security of participants into account due to the characteristic of vehicle fast mobility and the demand of driving security.First,in the process of data collection based on crowdsensing,the existing data collection strategies are based on traditional mobile crowdsensing,which can not be effectively applied to the vehicle environment.Second,contradiction between user privacy and data reliability is the main challenge faced with the crowdsensing.Because the perceived data is closely related to user identity to a certain extent,it is necessary to concern on this issue in the process of data collection.One feasible way is to do de-association.However,absolute deassociation leads to data untraceability,that is,malicious participants do not have to take responsibility for the data provided,which will bring about the risk of data unreliability.Thirdly,the implementation of crowdsensing can not be separated from the incentive mechanism for participants,while the existing researches have the problems of cumbersome process,long feedback time and user privacy disclosure.For the existing challenges mentioned above,this thesis mainly aims at data collection and incentive mechanism implementation of vehicle group perception,and provides corresponding privacy preserving solutions.Its main contributions are as follows:(1)In order to meet the common needs of data demanders and participating users in privacy protection and data validity,a heterogeneous two-tier fog sensing architecture is proposed for the first time in this paper.The lower tier mobile fog nodes are deployed on urban buses(called fog bus in the thesis),which can better improve the system responsiveness and shorten the data collection delay.Some data processing tasks are handed over to the FB to solve the resource consumption problem existing in centralized processing management,and the distributed architecture is used effectively to make the whole system more flexible.Moreover,the data encryption and aggregation strategy of FB effectively protects the privacy of participants and ensures the validity of data through the support of traceability strategy.In addition,in order to alleviate the storage pressure of third-party trust authority(TA),this work also proposes a traceability strategy based on data outsourcing,which enables TA to access data of traceability in a secure manner.(2)In order to make the crowdsensing implement the incentive mechanism more efficiently and simultaneously deal with the user privacy security problems caused by the introduction of incentive mechanism.On the basis of the proposed two-layer fog vehicle crowdsensing architecture mentioned above,combined with the bidding incentive model of all-payment,this work effectively improves the incentive feedback efficiency of the system.Then,utilizing a series of security authentication technologies,the user's privacy and security is protected during the process of rewards generation and the issuance.Finally,in order to ensure the reliability of perceived data,this work also introduces the reputation management mechanism and solves the problem of user privacy leakage respectively.