Visible to the public Improving Privacy-Preserving and Security for Decentralized Key-Policy Attributed-Based Encryption

TitleImproving Privacy-Preserving and Security for Decentralized Key-Policy Attributed-Based Encryption
Publication TypeJournal Article
Year of Publication2018
AuthorsZhang, Leyou, Liang, Pengfei, Mu, Yi
JournalIEEE Access
Volume6
Pagination12736—12745
ISSN2169-3536
Keywordsattributed-based encryption, authorisation, central authority, Collaboration, collusion attack, data privacy, DBDH assumption, Decentralized attribute-based encryption, decentralized key policy ABE schemes, decentralized KP ABE scheme, decentralized KP-ABE scheme, decisional bilinear Diffie-Hellman assumption, Encryption, Generators, key policy, key policy attribute based encryption, KP-ABE, policy-based governance, privacy, privacy-preserving decentralized KP-ABE, pubcrawl, Public key, public key cryptography, Resists, Scalability, user collusion avoidance
AbstractDecentralized attribute-based encryption (ABE) is an efficient and flexible multi-authority attribute-based encryption system, since it does not requires the central authority and does not need to cooperate among the authorities for creating public parameters. Unfortunately, recent works show that the reality of the privacy preserving and security in almost well-known decentralized key policy ABE (KP-ABE) schemes are doubtful. How to construct a decentralized KP-ABE with the privacy-preserving and user collusion avoidance is still a challenging problem. Most recently, Y. Rahulamathavam et al. proposed a decentralized KP ABE scheme to try avoiding user collusion and preserving the user's privacy. However, we exploit the vulnerability of their scheme in this paper at first and present a collusion attack on their decentralized KP-ABE scheme. The attack shows the user collusion cannot be avoided. Subsequently, a new privacy-preserving decentralized KP-ABE is proposed. The proposed scheme avoids the linear attacks at present and achieves the user collusion avoidance. We also show that the security of the proposed scheme is reduced to decisional bilinear Diffie-Hellman assumption. Finally, numerical experiments demonstrate the efficiency and validity of the proposed scheme.
DOI10.1109/ACCESS.2018.2810810
Citation Keyzhang_improving_2018