An Efficient and Privacy-Preserving Multi-User Multi-Keyword Search Scheme without Key Sharing

• Published in: ICC 2021 - IEEE International Conference on Communications pp. 1 - 6
• Main Authors: Song, Fuyuan, Qin, Zheng, Liang, Jinwen, Lin, Xiaodong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2021
• Subjects: Cloud computing; Conferences; Keyword search; multi-keyword search; Privacy; Privacy-preserving; Scalability; Search problems; searchable encryption
• ISSN: 1938-1883
• DOI: 10.1109/ICC42927.2021.9500478

Multi-keyword search, aiming to search the objects by a query request that consists of multiple keywords, has wide applications in personalized recommendation services. Mean-while, the fast development of cloud technology has given rise to a new trend that data are encrypted before being outsourced to a public cloud for users to enjoy pay-as-you-go services. However, most of the existing works primarily focus on the single keyword search, and consider a general scenario with a single owner and a single user. In this paper, we propose an efficient and Privacy-preserving Multi-user Multi-keyword Search (PMMS) scheme, which can support user scalability without key sharing. In particular, based on the matrix decomposition, a key derivation approach is integrated into our PMMS to generate secret keys and re-encryption keys. Furthermore, by employing threshold predicate encryption and leveraging the techniques of matrix transformation and proxy re-encryption, PMMS guarantees that only the comparison result of an inner product of two vectors and a pre-defined threshold is revealed, and enables the cloud server to perform multi-keyword search in an efficient and privacy-preserving manner. Security analysis shows that the confidentiality of owners' data and users' queries can be guaranteed. Extensive experiments on a real-world dataset demonstrate that PMMS is efficient in terms of multi-keyword search.