Privacy-preserving granular data retrieval indexes for outsourced cloud data

Abstract

Storage as a service has become an important paradigm in cloud computing for its great flexibility and economic savings. Since data owners no longer physically possess the storage of their data, it also brings many new challenges for data security and management. Several techniques have been investigated, including encryption, as well as fine-grained access control for enabling such services. However, these techniques just expresses the "Yes or No" problem, that is, whether the user has permissions to access the corresponding data. In this paper, we investigate the issue of how to provide different granular information views for different users. Our mechanism first constructs the relationship between the keywords and data files based on a Galois connection. And then we exploit data retrieval indexes with variable threshold, where granular data retrieval service can be supported by adjusting the threshold for different users. Moreover, to prevent privacy disclosure, we propose a differentially privacy release scheme based on the proposed index technique. We prove the privacy-preserving guarantee of the proposed mechanism, and the extensive experiments further demonstrate the validity of the proposed mechanism.