Consensus algorithms play a critical role in blockchains and directly impact their performance. During consensus processing, nodes need to validate and order the pending transactions into a new block, which requires verifying the application-specific data encapsulated within a transaction. This exposes the underlying data to the consensus nodes, presenting privacy concerns. Existing consensus algorithms focus on realizing application security and performance goals, but lack privacy-by-design properties or are resource-heavy and intended for securing permissionless blockchain networks. In this paper, we propose P-CFT, a zero-knowledge and crash fault tolerant consensus algorithm for permissioned blockchains. The proposed consensus algorithm provides inherent data privacy directly to the consensus layer, while still providing guarantees of crash fault tolerance. We conduct experiments using the Hyperledger Ursa cryptographic library, and the results show promise for integrating P-CFT into existing permissioned blockchain systems requiring privacy-preserving and crash fault tolerant features.

翻译：摘要：共识算法在区块链中扮演关键角色，并直接影响其性能。在共识处理过程中，节点需要验证待处理交易并将其排序生成新区块，这要求验证交易内封装的应用特定数据。该过程将底层数据暴露给共识节点，从而引发隐私问题。现有共识算法侧重于实现应用安全性与性能目标，但缺乏隐私设计特性，或资源消耗过高且专用于保障非许可区块链网络的安全性。本文提出P-CFT——一种面向许可区块链的零知识崩溃容错共识算法。该算法将数据隐私保护直接内置于共识层，同时仍能提供崩溃容错保障。我们基于Hyperledger Ursa密码学库开展实验，结果表明P-CFT有望集成至需要隐私保护与崩溃容错特性的现有许可区块链系统中。