The Cipher Text Retrieval Method And System Based On The Fully Homomorphic Encryption Technology

Aiming at the difficulty of the cipher text retrieval in the cloud server,this article makes a brief analysis and research on the current cipher text retrieval and fully homomorphic encryption.And a new method of cipher text retrieval based on the fully homomorphic encryption over integer is designed.Then a cipher text retrieval system based on this method is achieved.The main work is described as follows:(1)Document data processing stage: according to the TF-IDF vector model,the document data are calculated.The value of TF(Term Frequency)and IDF(Inverse Document Frequency)of the document data are obtained.Then a table of the encrypted word terms,TF,IDF and the ID of document is established.The index of document data is obtained.(2)Data encryption stage: RSA encryption key is used to encrypt the document,so as to realize the confidentiality of document data.The information of word terms in the plain index table are encrypted by using the hash encryption algorithm,and the confidentiality of word terms information is realized.The integer values of TF and IDF are encrypted by the fully homomorphic which is based on the method in this article.The encrypted documents,word terms,TF and IDF are uploaded to the cloud server,and then a cipher index is established at the cloud server side.The cipher index of the document data is obtained.(3)Retrieval service stage: the data users do the same hash encryption algorithm for the search keywords and upload them to cloud server.According to the hash value,the cloud server finds the encrypted TF value and IDF value.The value of TF multiplied by IDF is calculated based on the classical vector space model.The document score of the cipher text is obtained.And then the score is returned to the data owners to decrypt and sort,so as to get a higher score of the document data and get the required the ID of the document.The ID value of the document is returned to the data users.According to the returned ID,data users obtain the encrypted document data and use the RSA decryption key to decrypt.So the document is obtained and the the retrieval service is completed.Finally,in this design,the HBase distributed database is used to complete the experiment and the experimental data are analyzed.On the premise of guaranteeing the privacy of data,the cipher text retrieval operation is completed.