Secure And Efficient Dimension-reducing Ranked Query Method For Encrypted Cloud Data

Cloud storage technology provides a huge platform for users and enterprises.Users and enterprises can save local storage space,reduce costs and enjoy more convenient services by storing a large amount of data in the cloud.Therefore,cloud storage technology is widely used in various industries.However,if the user uploads the plaintext directly to the cloud server,it will cause information leakage,and the security of the user's personal information will not be guaranteed.At the same time,with the continuous expansion of the cloud storage information,the efficiency of data search will sharply decline,the cost of data maintenance will greatly increase,and the data security will not be effectively guaranteed.The traditional search methods can no longer meet the needs of users and enterprises Ask.Therefore,the research of secure and efficient sorting search scheme on encrypted data is extremely important.In view of data security and search efficiency,this paper studies as follows:First,we propose a secure and efficient linear dimension-reducing ranked query method for encrypted cloud data(DRQM_?).In order to improve the efficiency of data search,we use principal component analysis(PCA)algorithm to reduce the dimension of index matrix before encrypting the keyword index,so that the key dimension will also be reduced.When creating the trapdoor,we use the multiplication of query vector and eigenvector matrix from index matrix to reduce the dimensions,This will greatly improve the efficiency of data encryption and search efficiency.In view of the privacy violations of unauthorized users and untrusted servers,we adopt the reversible matrix encryption method to protect data information,and on this basis,we propose a random threshold setting method to further improve data security.In addition,we use the characteristics that query vector can be divided into multiple unit vectors and introduce unit matrix,so that the components of query vector also participate in the dimensionality reduction process,thus ensuring the consistency of dimensionality reduction of query vector and index matrix,not only improving the security,but also ensuring the accuracy of query.Secondly,although the DRQM_? scheme has greatly improved the query efficiency,the query accuracy of complex data relations is often difficult to meet the requirements.In order to ensure the query accuracy on the basis of improving the query efficiency,we propose a secure and efficient non-linear dimension-reducing ranked query method for encrypted cloud data DR QM_?.For non-linear data,we propose a novel fast non-linear dimension-reducing ranked query method with high security for encrypted cloud data(DRQM_?).For the no-linear and no-separable data,we use kernel principal component analysis algorithm(KPCA)to map the data into high space by nonlinear function,and transform it into linear separable data,and then reduce the dimension to decrease the component loss and further improve the query accuracyIn addition,this paper makes privacy analysis,performance analysis,accuracy analysis and security proof for the two schemes,and proves the feasibility,efficiency and security reliability of the scheme through theory and experiment.