Research On Privacy-preserving Query Mechanism For Spatio-temporal Data

With the popularization of location-aware devices,the application of query services based on spatio-temporal data provides convenience to people's daily life.In such query service applications,enterprise-based platforms collect spatio-temporal data provided by users,and provide users with various query services based on the collected data,such as KNN query,path matching service,range query service,etc.During the query process,the user needs to send their own location information and query conditions to the platform,and the platform returns the query result to the user according to the query content.With the substantial increase in the number of users,the cost of platform data storage and query computing also continues to grow.Cloud computing technology has become the optimal solution to deal with the scale of massive spatio-temporal data and the computational cost of processing user queries.Because the third-party cloud service providers that provide cloud computing cannot be completely trusted,in the process of storing spatio-temporal data and querying and computing based on location information,the platform must not only meet users' query needs,but also protect users' data privacy.For the commonly used KNN query and path matching services in spatio-temporal data,designing a spatio-temporal data query mechanism that takes into account query requirements and privacy protection is a current research hotspot.Many studies have proposed different solutions for these two query scenarios.But there are still problems with the proposed scheme.For KNN query scenarios,existing solutions cannot achieve a balance between security and efficiency.In addition,the existing solutions do not take into account the personalized needs of users in real scenarios.At the same time,in order to improve the accuracy of query results,the accurate evaluation of distance factors is an important indicator in KNN query,but the existing research is difficult to meet the needs of accurate evaluation of distance factors.For path matching query scenarios,existing solutions ignore the privacy protection issue.The efficiency of the existing solution will be reduced when faced with massive data,the flexibility of the solution is insufficient,and real-time path matching cannot be achieved.In order to solve the problems existing in the existing spatio-temporal dataoriented privacy-preserving query schemes,this dissertation constructs a spatiotemporal data-oriented privacy-preserving polynomial evaluation scheme and a privacy-preserving path matching query scheme.The innovative work of this dissertation mainly includes the following two aspects:(1)This dissertation designs a new encryption scheme to encrypt spatio-temporal data based on Paillier homomorphic encryption system and sequentially visible encryption scheme.The proposed virtual road network structure can effectively optimize the query efficiency based on the realization of spatio-temporal data storage.Finally,this dissertation uses a polynomial structure to respond to the individual needs of user query conditions and accurately evaluate distance factors.(2)This dissertation implements a secure query path matching method based on privacy-preserving set intersection operation.At the same time,improvements are made based on the original road network model to improve query accuracy and filter invalid operations.Finally,this dissertation tests and proves the efficiency of the proposed spatiotemporal data-oriented privacy-preserving query mechanism through comparative experiments.The comparative experimental results show that the proposed scheme can not only protect the privacy of users but also achieve efficient and personalized query in practical scenarios.