Research On Key Technologies Of Privacy Protection For Trajectory Data Release In VANETs

With the development of in-vehicle LBS applications and the rapid rise of mobile self-organizing network technologies,the topics related to VANETs have become hot topics in related fields at home and abroad,and the research on key technologies has received great attention from the industry.However,due to the large scale of VANETs,the openness of wireless channel communication and the predictability of the trajectory of vehicles,the data transmitted by each vehicle in VANETs can be obtained by means of eavesdropping or collusion with untrusted service providers.In the future,there will be more and more sensors in each vehicle in VANETs.These sensors will transmit a large amount of data to the background server every day,which contains the personal privacy information,behavior pattern,interests and other sensitive information of the vehicle and users.If the privacy information of any vehicle in VANETs is exposed,it will not be widely accepted by the public.Therefore,the premise of large-scale deployment of VANETs is how to effectively deal with the privacy security of the communication data between vehicles.Compared with other privacy data,the main privacy data in VANETs is the trajectory data of the vehicle.The trajectory data can be combined with the user's background knowledge to infer other privacy information of the user's vehicles.Attackers can analyze the privacy information,predict the trajectory of the vehicle and even trajectory the personal injury caused by the vehicle to the owner.This thesis systematically analyzes the privacy threats and privacy protection requirements in VANETs,and studies how to protect the trajectory privacy of vehicles in VANETs while ensuring the accuracy of LBS requests.The current mainstream trajectory privacy protection technologies are introduced.And compare our programs with existing technologies to illustrate the strengths and weaknesses of our programs and their main contributions:(1)For the trajectory privacy protection problem of location service request in interactive LBS application scenario in VANETs,this thesis firstly designs a new privacy protection mechanism by combining the difference privacy idea and the hidden area pseudonym replacement idea.This mechanism includes the stages of pseudonym replacement,noise addition and noise convergence.From the perspective of different users' demand for trajectory data,the trajectory granularity is stratified.While protecting the fine-grained trajectory of mobile users,certain coarse-grained trajectory is released to the service provider for value-added service research.The first stage applies the differential privacy index mechanism to the traditional pseudonym replacement mechanism to improve the data utility of the selected substitute.The second stage protects the fine-grained trajectory privacy by adding noise to the position coordinates on the fine-grained trajectory of the mobile user.The third stage converges the position points after noise addition through the convergence mechanism to further improve the data availability of these points on the trajectory.After these three stages,the service provider obtains a coarse-grained trajectory composed of the substitute id and the noised position of the real requesting user.Through theoretical analysis,these stages all satisfy ?-difference privacy;The feasibility of the trajectory privacy protection mechanism in this paper is judged by the experimental results.Finally,the availability and privacy of trajectory data can be adjusted by changing the privacy budget to achieve the goal of balancing privacy and availability.(2)This thesis studies the trajectory privacy problem caused by the non-interactive trajectory data release in VANETs,and proposes for the first time the trajectory data privacy protection mechanism combining differential privacy and anonymity in the environment of VANETs.The mechanism mainly includes four stages of space grid generation,trajectory segmentation,fragment equivalence class construction and trajectory publishing.In the first and second stages,according to the idea of divide and conquer,the long trajectory is processed anonymously in segments by means of grid division combined with the clue binary tree spatial index,so as to improve the data utility of trajectory data release.According to the different privacy requirements of different mobile users,personalized piecewise anonymous protection is carried out for the track fragments in the fragment equivalence class generated in the third stage.In combination with the idea of differential privacy and anonymity,the exponential mechanism is used to prevent the anonymous set from being attacked by secondary clustering.Finally,at the same level of privacy,the utility of the published trajectory data can be guaranteed by using the exponential mechanism to select a small number of obfuscated trajectory publications from the anonymous set.Through experiments on the trajectory privacy protection mechanism proposed in this paper and comparison with other schemes,the results show that the scheme proposed in this paper not only has a good effect on the trajectory privacy protection and service request quality of VANETs,but also can guarantee the operation performance of the mechanism.