Research And Implementation Of Homomorphic Encryption In Database

In recent years,data leakage problems have occurred frequently,and databases based on plaintext storage have great security risks.Encryption technology can ensure data security,but traditional encryption algorithms that do not contain any semantics need to trade data security at the expense of database computing power and efficiency.In order to ensure data security and improve database efficiency,ciphertext retrieval and computing technologies have been developed.Existing ciphertext retrieval schemes use traditional encryption algorithms to encrypt index fields,and data needs to be decrypted during retrieval,which is inefficient;in terms of ciphertext calculations,encrypted databases mostly use additive homomorphic encryption algorithms to achieve ciphertext additive homomorphism.It cannot meet other computing requirements of ciphertext in real scenarios,so ciphertext retrieval and ciphertext calculation technologies based on fully homomorphic encryption algorithms have emerged.However,the existing fully homomorphic encryption algorithms cannot be applied in practice due to the long time-consuming homomorphic calculation and the problem of too large key size.In response to the above problems,this thesis studies and designs a new PGSH homomorphic algorithm and ELGamal re-encryption algorithm,and applies it in the Postgre database to achieve database ciphertext retrieval and homomorphic computing functions.The work of this thesis mainly includes the following parts:(1)Aiming at the low efficiency of the existing ciphertext retrieval database,a PGSH homomorphic encryption algorithm based on the DGHV homomorphic algorithm is designed.The ciphertext retrieval algorithm in this algorithm only needs the public key to verify whether the two data match.At the same time,a ciphertext search database is constructed based on the PGSH homomorphic encryption algorithm.The database uses an encryption algorithm to encrypt the document index in the storage phase to generate a ciphertext index.In the retrieval phase,the ciphertext search algorithm is used to retrieve the ciphertext index.Compared with the existing ciphertext search database,the database in this thesis can implement retrieval based on the ciphertext index field without decrypting the private key,and has a higher retrieval efficiency.(2)Aiming at the problem that the fully homomorphic encryption algorithm cannot be applied in practice,an ELGamal re-encryption algorithm is designed.The re-encryption expression generation algorithm in this algorithm can directly perform arithmetic operations based on ciphertext.The homomorphic encryption database is constructed by combining the ELGamal re-encryption algorithm and the encrypted database,and the re-encryption expression generation algorithm is used to achieve the goal of ciphertext homomorphic calculation.Compared with the existing homomorphic encryption database,the database in this thesis can perform homomorphic addition,multiplication and mixed operations based on ciphertext,which is better suitable for real application scenarios.(3)In order to verify the validity of the homomorphic encryption algorithm proposed in this thesis,the PGSH algorithm and the ELGamal re-encryption algorithm are applied in the Postgre database to implement an encrypted database with the functions of ciphertext retrieval and homomorphic calculation.Compared with existing encrypted databases,the homomorphic encryption algorithm proposed in this thesis can directly retrieve and calculate based on ciphertext,ensuring data security and having higher efficiency.