University of Wollongong
Browse

Threshold Ring Signatures with Accountability

journal contribution
posted on 2024-11-17, 15:00 authored by Xuan Thanh Khuc, Willy Susilo, Dung Hoang Duong, Fuchun Guo, Kazuhide Fukushima, Shinsaku Kiyomoto
Threshold ring signatures (TRS) allow several signers to sign the same message on behalf of a group. This scheme is fully anonymity in that a signature reveals the number of signers who created the signature but tells nothing about the identity of all signers who participated in generating the signature. However, in several scenarios, it is required that signers should take responsibility for their actions and hence traceability of the signer’s identity is needed. In this paper, we introduce a new type of threshold ring signature, called threshold ring signature with accountability (ATRS). It is a balance of anonymity and accountability. In addition to achieving privacy as in the TRS scheme, an ATRS scheme also allows a designated opener to learn the identity of all signers who participated in creating the signature when necessary. Notably, the signers can decide whether they want to disclose their identity if they have any concerns about the designated opener. Moreover, the opener must provide a verifiable opening to ensure that he/she cannot hold a signer responsibility for a signature that was not created by him/her. Our new type of threshold ring signature combines the notion of tracing in traditional anonymity-oriented signature primitives with threshold ring signatures and can enable various new and appealing privacy-preserving applications. We formalise the definition and security requirements for ATRS. We present a generic construction to demonstrate the feasibility of designing ATRS in a modular manner from commonly used cryptographic building blocks (verifiable random function, public-key encryption, pre-constrained encryption and NIWI). Our scheme is secure in the plain model that does not require the heuristic of random oracle or trusted setup assumptions.

Funding

University of Wollongong (LP190100984)

History

Journal title

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume

14895 LNCS

Pagination

368-388

Language

English

Usage metrics

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC