Secure and Efficient Data Processing for Cloud Computing with Fine-Grained Access Control

Abstract

Nowadays, with rapid development of cloud computing, many information systems are running on the cloud platform. However, the cloud servers are not fully trustworthy, for the purpose of privacy preserving, the users need to encrypt their data before uploading it to the cloud. However, this also brings challenges in utilizing the data. Generally speaking, several desirable properties should be achieved for data processing on the cloud platform. First, the cloud servers should be able to perform computations on the encrypted data without learning users’ sensitive information. Second, fine-grained access control should be enforced on the computed results. Third, flexibility requires that the identities who can access the computed results should be unknown when these results are generated. Fourth, the scheme should have low overheads on computation and communication. To the best of our knowledge, most of the existing schemes cannot satisfy these requirements simultaneously. In order to address this issue, we propose a secure and efficient privacy preserving data processing scheme for cloud computing with fine-grained access control, using a homomorphic proxy re-encryption scheme and an efficient attribute-based encryption scheme. Security analyses prove that it satisfies all the desirable security properties, and performance evaluation demonstrates that it is more efficient than the state-of-the-art schemes targeting similar problems. In particular, the size of ciphertexts and the decryption time for the computed results are constant in our scheme, regardless the access structure. Therefore, our scheme contributes to a more practical data processing scheme for the cloud platform with fine-grained access control.