The Research On Data Encryption And Ciphertext Search Algorithm In Cloud Storage

With the development of cloud computing technology,cloud storage technology has been a large number of users with its convenience,low price and high scalability.However,in recent years,with the large number of cloud service providers(Cloud Service Provider,CSP)occur data leakage events,people have to consider the security issues of the data in the cloud.To protect the data stored in the cloud,most users choose to encrypt the data before uploading to the server.However,data encryption not only protects data security,but also brings the problem of difficult query and data sharing.Therefore,how to solve the security of storage data,and how to fast and secured retrieve the encrypted file,are the key issues which need to solve on security research areas in the cloud storage.In view of this,this article will deeply study in the ciphertext access control,and ciphertext retrieval,the paper's research mainly includes the following aspects:Firstly,consider to the cloud storage having volume file data and the data sharing flexible,this paper studies the attribute-based encryption(ABE)and the attribute encryption method(Multi-Authority ABE)in the multi-authority environment.After analysis of the cloud data security and sharing,this paper chooses MA-ABE as the research object of this thesis.Second,when studying MA-ABE algorithm,we find that in the current algorithm it exits these problems,such as excessive dependence on Central Authority(CA),weakly resistance to collusion attack.For this problem,this paper proposes an access control scheme based on MA-ABE,which divides the key into two parts:user's private key and attribute private key.Which are generated and distributed by Data Owner(DO)and Attribute Authority(AA).Without CA,it can be avoid the problem of key leakage caused by joint attack between AA.The security analysis proves that the scheme can not only resist the joint attack between AA,but also has ability to resist the plaintext attack and selection attack.In addition,for the retrieve slowly and difficult to support multi-keyword fuzzy query problem,this paper proposes a new retrieval scheme,which is the ciphertext retrieval scheme based on curve fitting.In the scheme,the ciphertext is mapped into a curve by the index algorithm,and the keyword is mapped into the coefficients in the curve.With the idea of curve fitting,the ciphertext matching is transformed into the matching of the curves.This scheme can support the query of fuzzy keywords by improving the index algorithm.In addition,this scheme supports search scopes defined by user,and the search results can be return from exact match to approximate match based on this search range.The experimental results confirmed that this scheme has a better performance in the retrieval efficiency and multi-keyword fuzzy searching.