Visible to the public A Verifiable Dynamic Multi-user Searchable Encryption Scheme without Trusted Third Parties

TitleA Verifiable Dynamic Multi-user Searchable Encryption Scheme without Trusted Third Parties
Publication TypeConference Paper
Year of Publication2019
AuthorsYe, Fanghan, Dong, Xiaolei, Shen, Jiachen, Cao, Zhenfu, Zhao, Wenhua
Conference Name2019 IEEE 25th International Conference on Parallel and Distributed Systems (ICPADS)
Keywordsauthorisation, cloud computing, cloud server, composability, cryptography, dynamic management, encrypted data, Multi-user, multiuser searchable encryption schemes, multiuser settings, pubcrawl, public key cryptography, resilience, Resiliency, Searchable encryption, secure data sharing, trusted third party, user access permission matrices, user permission, user search permissions, verifiable dynamic multiuser searchable encryption scheme, verifiable multiuser searchable encryption schemes, verification, weak trust
AbstractSearchable encryption is a cryptographic primitive that allows users to search for keywords on encrypted data. It allows users to search in archives stored on cloud servers. Among searchable encryption schemes, those supporting multiuser settings are more suitable for daily application scenarios and more practical. However, since the cloud server is semi-trusted, the result set returned by the server is undefined, and most existing multi-user searchable encryption schemes rely heavily on trusted third parties to manage user permission. To address these problems, verifiable multi-user searchable encryption schemes with dynamic management of user search permissions, weak trust on trusted third parties and are desirable. In this paper, we propose such a scheme. Our scheme manages user permission and key distribution without a trusted third party. User search permission and user access permission matrices are generated separately to manage user permissions dynamically. In addition, our scheme can verify the result set returned by the cloud server. We also show that our scheme is index and trapdoor indistinguishable under chosen keyword attacks in the random oracle model. Finally, a detailed comparison experiment is made by using the actual document data set, and the results show that our scheme is efficient and practical.
Citation Keyye_verifiable_2019