The Research And Implementation Of Algorithm For Cloud Storage Database Ciphertext Retrieval

With the quiet coming of the "cloud" era,individuals and enterprises are more and more inclined to store the required data in cloud storage databases.However,the security of cloud storage data has gradually increased,and the security of database data has attracted wide attention.In this context,in order to ensure the security of cloud sensitive data,the conventional approach is to encrypt cloud data to ensure data security.But while database encryption guarantees security,it also brings problems of data retrieval and computation.In database mode,relational database has been in the leading position in the market for its advantages of supporting complex condition query,simple operation and perfect data structure.For encrypted relational database,its data no longer have the original characteristics of ordering and calculability.At the same time,many computing functions in relational database can not play a role in ciphertext environment.Computing and retrieval are the indispensable core functions of database.Therefore,how to efficiently implement ciphertext retrieval and computation in relational database while ensuring data security is of great significance.Based on CryptDB ciphertext retrieval scheme,this paper optimizes the retrieval scheme and extends its functions,including the following work and innovations:1.A new f-mOPE ordered encryption scheme based on mOPE ordered encryption scheme.In CryptDB ciphertext retrieval scheme,ordered encryption algorithm mOPE is used for ordered encryption calculation.In this paper,an improved f-mOPE ordered encryption scheme is used to insert and find elements,which reduces the number of interaction between client and server to log(n)(where n is the total number of elements)to a maximum of alpha times,where alpha is constant.This paper proposes a ordered encoding block governance scheme to reduce the total data.The effect of number on reordering of preserved coding in the scheme.With the increase of the total number of data n,the number of reordering operations of the balanced binary tree maintained on the server side in mOPE also increases with the increase of n.In this paper,the reordering operation number is controlled at a constant level by using the block governance scheme of preserving order coding,which is not affected by the total number of data n.This scheme adopts the strategy of space-for-time to determine the order based on the ordered index.Reduce thecommunication overhead of the whole sequential encryption scheme,and reduce the performance loss of the server side.In terms of security,this paper proposes a fuzzy encryption algorithm and a partial homomorphic encryption algorithm.On the basis of ensuring data privacy,this algorithm can not only resist statistical attacks,but also effectively reduce the computing overhead of servers and improve the efficiency of database processing.2.Attribute-based access control for ciphertext retrieval.In cloud storage environment,stored data is usually shared by many parties,and different users have access to part of the data.In order to prevent data privacy leakage and ensure data security in cloud storage,this paper proposes a fine-grained attribute-based access control model,which combines attribute encryption with onion encryption model.As the outermost layer of onion encryption layer,it can accurately control users' access rights to data columns,and at the same time,it can greatly reduce the computational overhead caused by privilege revocation.3.By introducing f-mOPE ordered encryption scheme into CryptDB and using attribute-based access control model,the experimental analysis shows that the scheme has higher security and efficiency than CryptDB,and has certain feasibility and practical value.