Privacy-preserving auditing scheme for shared data in public clouds

Abstract

Recent advances in cloud storage have enabled users to outsource large amounts of data to a remote cloud server in order to reduce storage and management costs, and share files among many users in a group. However, how to efficiently audit the integrity of shared data while maintaining data privacy and user identity anonymity, is still a critical issue. We propose a novel public auditing scheme for data stored in a remote cloud server and shared among users in a large group. In particular, the proposed scheme incorporates group signature, homomorphic message authentication code to create data block tags, so that it can support public auditing and provide user identity anonymity. Furthermore, we use the random masking technique in the proposed scheme to preserve data privacy from the third-party auditor. The correctness and security analyses demonstrate that the proposed scheme is correct and provably secure under a robust security model. The performance evaluation and experimental results show that the proposed scheme is efficient while maintaining the desirable security properties.