Research On Trajectory Flow Data Publishing Method Based On Differential Privacy

With the development of the Internet and location-aware devices,Location Based Services(LBS)has been widely used to provide people with various convenient Services.When people use LBS,they usually need to submit query requests to the LBS service provider,which may carry other sensitive information of the user.Therefore,trajectory privacy protection in location service has become the focus of attention and research hotspot.At present,the existing trajectory privacy protection methods have some shortcomings,such as the inability to effectively resist background knowledge attacks and the poor availability of published data.In this paper,a systematic and in-depth study is carried out on the release of trajectory flow data in location service.According to different scenarios,a trajectory flow data release method based on differential privacy(DP)is proposed,which can not only protect the privacy security of users,but also ensure the availability of data.The main research contents are as follows:(1)In the road network space composed of simple roads(e.g.,highways),a trajectory flow data publishing method based on differential privacy(RS-DP)is proposed.Firstly,the topology of the road network is represented by a directed graph according to the correlation between the nodes of the road network.Then,based on the topology of the road network and the user trajectory data,the original trajectory flow data is generated.Laplace mechanism is used to add noise to the original trajectory flow to realize differential privacy protection.Finally,in order to improve the availability of data,the post-constraint algorithm(Con Ad-GH)is used to optimize the trajectory flow containing noise.Finally,based on both synthetic datasets,we do experiments to evaluate the performance of the proposed method.The results illustrate that RSDP can reduce errors and provide better quality of service.(2)In the free space composed of irregular and complex roads(e.g.,shopping malls),a trajectory flow data publishing method based on differential privacy(FS-DP)is proposed.Firstly,the quadtree structure is used to recursively count the spatial regions in the free space,and the differential privacy budget is divided according to the hierarchy.Then,the Laplace noise is added to the flow value in each region to realize differential privacy protection.Finally,according to the consistency constraints of distance,direction and capacity,the post-constraint algorithm(Con Ad-QT)is used to optimize the trajectory counting including noise.The validity of the scheme was verified by real data set experiment.The experiment results show that the proposed method(FS-DP),compared with other methods,can reduce the average relative error and improve the usability of query results.