Cloud Data Storage Model Based On Homomorphic Encryption Technology

Cloud Storage is a new storage architecture and service developed from cloud computing, and it has some advantages such as low equipment access and low cost use. But the cloud storage security has been an important issue which restricts cloud storage service to spread for use. Therefore, research institutions and manufacturers have introduced cloud security storage solutions.In this thesis, the cloud security storage model based on homomorphic encryption technology, integrated Gentry full homomorphic encryption scheme and BGV homomorphic algorithm, and optimize them to achieve ciphertext data secure storage and use convenient with its homomorphic property. By improving the ciphertext retrieval, data authorization, share and update of ciphertext mechanism to ensure that the data use security and convenience under non-decrypt environment. Ciphertext retrieval integrated linear scan algorithm and inverted index structure to establish index document dynamic formation mechanism. And the introduction of TF-IDF can improve the accuracy of ciphertext retrieval.Based on the study of the existing models of cloud security storage, we put a evaluation model based on fuzzy comprehensive evaluation method to realize the compare of cloud storage models. The evaluation model use the experts evaluation、membership function and weight vector to evaluate the performance of cloud storage model.In this thesis, the design and implementation of encryption algorithem test 、ciphertext retrieval test and storage model evaluation test verify the rationality and feasibility of the theory in before.