Nearest Neighbor Query Method Research For Cloud Based On Road Network Data

In recent years,due to the rapid development of mobile Internet,spatial data services represented by queries in European space and road network space have been used widely.While bringing convenience to users,there are hidden dangers in information security.While implementing spatial data query,it can also protect the privacy of user information.Such research has become a hot spot in the field of spatial database.This paper take large-scale spatial network data as research object,and proposes secure data nearest neighbor(SDNN)query method for spatial network data,which solves many problems that need to ensure information security in real life neighbor query problem applications.The secure data nearest neighbor query of spatial road network data is a combination of cloud storage technology and geographic location-based services.First,data encryption is used to ensure data security,and a hybrid encryption method is proposed for cloud road network data.Combining DES and AES encryption algorithms,the DES encryption key is used to encrypt the data content as an intermediate ciphertext,and then the cipher-text is encrypted by the AES encryption key as the final ciphertext to achieve road network data content protection.According to the position information of interest points in space,the Voronoi diagram is used to construct a polygon block and the space is divided into multiple adjacent spatial units.The space midpoint information is replaced with block information to achieve road network data position information protection.The spatial unit can be divided into multiple subregions based on the spatial unit.After receiving the request from the user to issue a query,the part half cut method(PHC),the iterative division between points method(IBP),and the half split method(HM)are proposed to divide the entire space area according to the attributes of the interest point,and then using structured way to create a memory index by Quad-tree and R-tree,then realize nearest neighbor query.In order to improve the query efficiency,the memory hash table for establishing space units and partitions is proposed,and a fast search is performed by directly searching the storage locations of the space units.Through experimental tests Open Street Map Project on real data sets,we analysis and compare their performance and efficiency.Experiments show that this method can implement the nearest neighbor query in the road network safely and effectively,which can keep the query performance,and guarantee the security of the data.