Study On Key Technologies Of Spatial Encryption Database

With the widespread use of location-aware smart devices in recent years,data containing geospatial information has grown at an unprecedented rate,and the era of spatial big data has come.Spatial data has typical big data characteristics such as large data volume and many types of data.In order to reduce their data management costs,small and medium-sized enterprises often choose to host spatial big data on a third-party cloud platform.In order to prevent the cloud storage service provider from leaking data,the spatial data can be stored on the cloud platform based on the encrypted database,and the query operation can be performed directly on the encrypted data.Only the data hosting user can decrypt the query result.However,the traditional encrypted database represented by CryptDB currently only supports operations on relational encrypted data,and does not support operations such as spatial query on spatial encrypted data.In view of the above problems,this paper improves on CryptDB to support spatial data types,and implements spatial operations such as storage,indexing,and querying of spatially encrypted data.It is further refined into a more complete spatially encrypted database prototype system.The main work of this paper is as follows:(1)Design and implementation of a spatially encrypted data range query scheme.This paper implements two spatial encryption data range query schemes:Geohash-OPE and FastGeo.The Geohash-OPE scheme is designed by this paper.It first uses geohash coding to establish spatial index on the spatial point data,then encrypts the geohash value through the order-preserving encryption,and finally implements range query based on the encrypted geohash value;The FastGeo scheme refers to the design idea in paper^[5].Firstly,AES-CBC-256 and SSW algorithm are used to encrypt the spatial point data respectively,and then the encrypted spatial point data is organized into a secondary index similar to the hash table.Through this hash table,the number of points to be screened during the query is reduced,and the range query efficiency is improved.(2)The design and implementation of KNN query scheme for spatial encrypted data.This paper also implements two schemes of KNN query for spatial encrypted data.The two KNN query algorithms convert KNN query into range query through circular expansion at the initial execution,and then use the above two different range query schemes to realize KNN query function.(3)Design and implementation of a spatially encrypted data query algorithm scheme in CryptDB.Based on the original functions of CryptDB,the two range query and KNN query solutions implemented in this paper can be integrated into CryptDB,so that the application layer can easily store and query the spatially encrypted data in CryptDB..Finally,this paper builds a prototype system in the virtual machine,uses the random spatial point data set to test the performance of spatial data entry,spatial encryption point data range query scheme and KNN query scheme,and compares the experimental results with My SQL's experimental results under the plaintext spatial point data set,which confirmed the practicability of the prototype system implemented in this paper.