A lattice-based privacy-preserving decentralized multi-party payment scheme

Jisheng Dong, Qingni Shen, Junkai Liang, Cong Li, Xinyu Feng, Yuejian Fang

Abstract: The use of cryptocurrencies has become an emerging and popular way of trading as they gain legitimacy. To address the issue of privacy leakage, some techniques to hide transaction amounts have been proposed such as the MimbleWimble protocol. However, these privacy enhancement schemes basically apply to one-to-one tradings between one payer and one payee, resulting in cryptocurrencies not being used in broader scenarios such as more than one payer or payee (referred to as multi-party transactions in this paper). In this work, we propose a new privacy-preserving decentralized multi-party payment (PDMP) scheme that ensures the transaction amounts in multi-party transactions remain confidential to other parties, and define the ideal functionality for it which captures the privacy and security properties in cryptocurrencies. Then we instantiate a lattice-based PDMP protocol in a hybrid model which can universally composable (UC) securely realize the functionality with a simulation-based security proof. We construct a lattice-based verifiable multi-secret sharing scheme and a lattice-based multi-prover non-interactive zero-knowledge argument to support the protocol, both of which enjoy the security in the future quantum computer era. At last, we have carried out experimental implementation of the scheme to prove its feasibility.