Research On Secure And Advanced Search Over Encrypted Cloud Data

Recently,cloud computing has becoming more and more popular.There are increasingly number of people outsourcing their data into cloud.However,the data stored in cloud contains lots of sensitive information,which can bring privacy risk.On the one hand,the cloud server may mine the outsourced data for commercial use.On the other hand,the vulnerability cloud storage system leads to frequent privacy leakage.Therefore,Encrypting the sensitive data before outsourcing is generally regarded as the most efficient way to conserve privacy.Nevertheless,the losing of data usability after encryption makes it very difficult to implement data search function over encrypted data.The existing search schemes mainly focus on some basic functions,such as boolean search and similarity search.They cannot properly support advanced search function.It is a very challenging task to conduct advanced search type on encrypted data.To tackle this problem,the main contributions of this thesis are as follows:1)A new searchable encryption scheme that realizes structured queries with generalized pattern matching over encrypted data.It allows users to conduct generalized pattern matching queries on textual attribute values of structured data,and joint them with logical operators(AND,OR,NOT)to search over multiple attributes of the data sets.In the proposed scheme,the comparison of strings is transformed into computation of fingerprints vectors.The results which are matched exactly can be selected by the threshold-based pattern matching algorithm.Moreover,an attribute-binding method is used in the scheme to protect the search pattern of users to attributes.Security analysis proves that the scheme is KPA-secure.The experiment results show that the scheme achieves high search accuracy and practical search efficiency.2)A secure searchable encryption scheme that realizes relevance ranking for multikeyword search over cloud data.There have been some existing searchable encryption schemes that support result ranking.Among these schemes,the relevance of index string and search query is the most wildly used measurement when ranking the search results.However,the existing schemes mainly use the TF-IDF weight to calculate the relevance value.They have not comprehensively considered other factors which may have effect on the relevance.To solve this issue,a secure searchable encryption scheme that realizes relevance ranking for multi-keyword search over cloud data is proposed in this thesis.The scheme uses a new relevance computing method that combines TF-IDF and word cooccurrence factor.Security analysis and experiment results demonstrate that the proposed scheme is secure and efficient.In conclusion,this thesis aims at designing secure and efficient advanced search scheme in cloud environment and summarizes the related works of searchable encryption schemes.A structured search scheme and a multi-keyword ranked search scheme over encrypted cloud data are proposed in this essay.The proposed schemes extend the functionality of searchable encryption and insure the security of data meanwhile.Therefore,this thesis has both theory significance and application value.