Enabling Secure And Efficient Multi-keyword Fuzzy Retrieval Over Encrypted Data

With the advent of encrypted cloud,the information security problem is highlighted gradually on cloud.In order to protect the security of sensitive data,it firstly encrypts their data on client before uploading the ciphertext to the server.Obviously,the encrypted data inevitably lose some characteristics,which makes it more difficult for keywords retrieve.Therefore,it becomes a challenging problem to finish flexible keywords retrieval on encrypted data.Though many solutions have been proposed,they can't solve the inadequates of fuzzy retrieval and trapdoor linkability effectively.In addition,many schemes adopted the inverted index,which can not resist mass quantity data.It also reduce the retrieval efficiency and space utilization.In a word,how to realize the security of the cipher text retrieval at the same time,improve the efficiency and accuracy of keywords fuzzy retrieval become a difficulty and a focus of study.In the research,the existing multi-keyword schemes for ciphertext retrieval can't achieve efficient fuzzy retrieval and can't resist different types of attacks.Under the current environment of secure cloud and other retrieval schemes,this paper studies multiple keywords and propose a fuzzy retrieval method for ciphertext(Multi-Keyword Fuzzy Query,MKFQ).Moreover,with encrypted conditions,the MKFQ can also resist the trapdoor correlation attack,adaptive selection keywords attack(IND-CKA)and similarity attack.The research of this topic will be realized to a certain extent,under the condition of ciphertext and multi keyword fuzzy search,and to ensure the security and effectiveness of ciphertext retrieval.The specific work of this paper is as follows:1.In this paper,a new full-text keywords index structure is designed,and a multi-keyword fuzzy retrieval algorithm is implemented to improve the standard Bloom Filter,which is based on the advantages of the current keywords retrieval scheme on encrypted data.Experiments show that the scheme not only supports the personalized retrieval of users,but also improves the system space utilization.2.Based on the matrix theory,the MKFQ method is implemented by the symmetric encryption algorithm,and the performance of the MKFQ method is analyzed from the aspects of security and validity according to different evaluation indexes and attack methods.The performance of the MRSE-1,MRSE-2 and RQED methods are used to comparing and analyzing,and the correctness of the algorithm is verified by the comparison of the results of the Multi-Keyword Ranked Search method.