Paper 2020/998 Original: eprint.iacr.org/2020/998 ↗

Lightweight Virtual Payment Channels

Abstract

Blockchain systems have severe scalability limitations e.g., long confirmation delays. Layer-2 protocols are designed to address such limitations. The most prominent class of such protocols are payment channel networks e.g., the Lightning Network for Bitcoin where pairs of participants create channels that can be concatenated into networks. These allow payments across the network without interaction with the blockchain. A drawback is that all intermediary nodes within a payment path must be online. Virtual Channels, as recently proposed by Dziembowski et al. (CCS'18), allow payments without this limitation. However, these can only be implemented on blockchains with smart contract capability therefore limiting its applicability. Our work proposes the notion of --Lightweight-- Virtual Payment Channels, i.e. only requiring timelocks and multisignatures, enabling Virtual Channels on a larger range of blockchain systems of which a prime example is Bitcoin. More concretely, other contributions of this work are (1) to introduce a fully-fledged formalization of our construction, and (2) to present a simulation based proof of security in Canetti's UC Framework.