Research On Data Retrieval Technology For Suppressing Information Disclosure In Cloud Environment

In the Internet age,the large amount of data generated due to the overload of multimedia information cannot be properly managed.Since the cloud has the advantages of convenient data management and low price,data owners prefer to outsource operations such as the storage of massive data and related calculations to the cloud for processing.However,while providing convenience,cloud servers also bring about the problem of information leakage and abuse.The users outsource data to the cloud environment and lose their ability to control data.It may lead to the leakage and abuse of private data such as personal information,which violates the original intention of using cloud servers.Although the data can be encrypted and uploaded,it also hinders the user’s flexible access to the encrypted data.The Searchable Encryption(SE)technology has solved the problem of retrieval in the ciphertext domain,but most of them improve the search efficiency at the cost of leaking information.Information leakage seriously threatens the security of data,so how to retrieve data efficiently and safely has become a problem that has to be solved.This thesis puts forward the following solutions to solve the above problems:1.For the index structure,this thesis proposes a data retrieval scheme that suppresses information leakage of the index.Currently,there are few searchable encryption schemes that hide both search pattern and access pattern on index structures.To solve the adversary’s data recovery attacks that utilize the leakages of search pattern and access pattern,this thesis proposes to use the 1-out-of-n oblivious transfer(OT_n~1)protocol and Paillier encryption to build an efficient and secure data retrieval model.Among them,the OT_n~1 protocol is used to obfuscate search objects while protecting the client and server privacy.The Paillier encryption is used to implement a shuffling algorithm to change the access path while hiding the search pattern and access pattern on the index structure.2.For the file set,this thesis proposes a data retrieval scheme to suppress information leakage of file set.The current searchable encryption schemes suffer from response length leakage.In order to solve the adversary’s data recovery attack by utilizing response length leakage,this thesis proposes to use the Differential Privacy(DP)strategy based on random response to hide the response length of the file set.Privacy processing of data locally before users upload sensitive data to protect the privacy of user data without the need for trusted third parties.Finally,this thesis proves the security of the proposed scheme,and uses online public data sets to experimentally verify the proposed scheme.Theoretical analysis and experimental results show that the two schemes proposed in this article are safe and feasible.