CNS Core: Small: Network Architecture and Routing Protocols for Payment Channel Networks

CNS 核心：小型：支付通道网络的网络架构和路由协议

• 批准号: 1910676
• 负责人: Mohammad Alizadeh
• 金额: $ 50万
• 依托单位: Massachusetts Institute of Technology
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1910676&HistoricalAwards=false
• 关键词: CNS; Core; Small; Network; Architecture

Blockchain technology is attracting considerable interest for its potential to enable new decentralized, secure applications such as payment systems. However, current blockchain systems suffer from severe scalability limitations. For example, blockchain-based digital currencies have orders of magnitude worse transaction throughput and confirmation latency than the established payment systems we use in our daily lives. This project proposes to tackle this fundamental scalability challenge by investigating so-called "payment channel networks". A payment channel network enables participants to transfer funds securely through a set of intermediaries, without recording every transaction on a public ledger (the blockchain). This project starts with the observation that payment channel networks have similarities to data communication networks (and also important differences), but their designs are primitive by comparison. The goal this research is to develop network protocols for payment channel networks --- protocols that are tailor-made for these networks while being deeply rooted in decades of research on data networks. If successful, this project would enable principled, efficient, and cost-effective approaches to building decentralized payment systems and improving the scalability of many blockchain systems. A central challenge for a payment channel network is how to route payments across the network efficiently. Routing payments is non-trivial since it requires finding paths with sufficient funds and keeping payment channel funds balanced. Payment channels used more frequently in one direction eventually run out of funds in that direction blocking further transactions through them. Inefficient routing limits network throughput and increases transaction costs, as it forces routers to deposit a large amount of funds into their payment channels up front, or pay high transaction fees to repeatedly rebalance payment channels on the blockchain. This research seeks to develop a new payment channel network that achieves high transaction throughput at low cost. To this end, the researchers plan to (1) develop a packet-switched payment channel network architecture, (2) invent imbalance-aware routing and rate control algorithms with theoretical guarantees, and (3) investigate privacy and incentive-compatibility of payment channel network protocols.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

区块链技术因其在实现支付系统等新的去中心化、安全应用程序方面的潜力而引起了极大的兴趣。然而，当前的区块链系统存在严重的可扩展性限制。例如，基于区块链的数字货币的交易吞吐量和确认延迟比我们日常生活中使用的现有支付系统要差几个数量级。 该项目建议通过研究所谓的“支付通道网络”来解决这一基本的可扩展性挑战。 支付通道网络使参与者能够通过一组中介机构安全地转移资金，而无需在公共分类账（区块链）上记录每笔交易。 该项目首先观察到支付通道网络与数据通信网络有相似之处（也有重要的区别），但相比之下，它们的设计很原始。这项研究的目标是开发支付通道网络的网络协议——为这些网络量身定制的协议，同时深深植根于数十年的数据网络研究。如果成功，该项目将为构建去中心化支付系统并提高许多区块链系统的可扩展性提供有原则、高效且具有成本效益的方法。 支付渠道网络的一个核心挑战是如何在网络上有效地路由支付。路由支付并非易事，因为它需要找到资金充足的路径并保持支付渠道资金平衡。在一个方向上更频繁使用的支付通道最终会耗尽该方向上的资金，从而阻止通过它们进行进一步的交易。低效的路由限制了网络吞吐量并增加了交易成本，因为它迫使路由器预先将大量资金存入其支付通道，或者支付高额交易费用以反复重新平衡区块链上的支付通道。这项研究旨在开发一种新的支付渠道网络，以低成本实现高交易吞吐量。为此，研究人员计划（1）开发分组交换支付通道网络架构，（2）发明具有理论保证的不平衡感知路由和速率控制算法，以及（3）研究支付通道的隐私和激励兼容性该奖项反映了 NSF 的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

High Throughput Cryptocurrency Routing in Payment Channel Networks

• 发表时间: 2018-09
• 作者: Vibhaalakshmi Sivaraman;S. Venkatakrishnan;Kathleen Ruan;Parimarjan Negi;Lei Yang;R. Mittal;G. Fanti-

DispersedLedger: High-Throughput Byzantine Consensus on Variable Bandwidth Networks

DispersedLedger：可变带宽网络上的高吞吐量拜占庭共识

• 发表时间: 2022
• 期刊: 9th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22
• 作者: Yang, L;Park, S;Alizadeh, M;Kannan S;Tse, D.
• 通讯作者: Tse, D.