Research On Privacy-Preserving Multi-keyword Search Algorithm

Cloud computing provides cheap computing resource and storage space for customers,hence,many companies and individuals willing to deploy their data and procedures in cloud servers.However,the cloud server providers may analyze these data illegally.Therefore,when we enjoy the conveniences of cloud computing,we should to take the privacy of our data into account.For protecting privacy,a common-used method is that encrypting whole data before outsourcing to the cloud servers.However,encryption will reduces the availability of data,and it makes the traditional information retrieval methods cannot be directly adopted to encrypted data.Hence,searchable encryption has become a hot research field,which enables cloud servers query processing over encrypted data and privacy-preserving simultaneously.Although many mature methods have been proposed to support different kinds of searches,these methods cannot hide the access pattern of query processing,and most of them are inefficiency.On the one hand,for improving the efficiency of query processing,we propose a group top-k search scheme(GMTS),which based on the divide-and-conquer algorithm,and it can improve the efficiency of the index construction,trapdoor generation and query processing of searchable encryption.On the other hand,in order to prevent privacy leaks,we propose a random traversal algorithm(RTRA),which makes the cloud server sends back different results and visited different paths on index for the same query,and at the same time,it maintains the query accuracy unchanged.We also propose an efficient and secure random group top-k search scheme(RGMTS),which absorbs the advantages of RTRA and GMTS.Finally,through sufficient experiments and security analysis,we prove our proposed methods are more efficient in index construction,trapdoor generation and query processing than the existing methods,and it is more secure.