Practical Byzantine improved algorithm based on node-independent validation
The PBFT consensus algorithm solves the Byzantine node tolerance problem of Raft and Paxos consensus algorithms, but the three-stage confirmation leads to its poor scalability and high communication latency . In response to above problems, this paper proposes an improved PBFT algorithm based on node-independent validation, briefly as V-PBFT. V-PBFT divides nodes into a double layer structure with multiple leaders, and sets up global validation nodes, intra-group followers, suspicious nodes and malicious nodes. Firstly, V-PBFT adopts the PBFT protocol to reach consensus among leaders. Validation nodes guarantee the consistency of the state of leaders. Then, leaders broadcast I n t r a − c o n s e n s u s messages to intra-group followers. Secondly, validation nodes broadcast V e r i f y messages to all nodes except malicious nodes and leaders. The messages from the leader and validation nodes are mutually verified, enabling followers to judge the correctness of client requests. Finally, the credit value is set for followers to reduce the probability on malicious view-changes. Experimental results show that V-PBFT reduces the communication complexity of PBFT from the quadratic level to the linear level and thus its communication latency is significantly reduced. • The communication complexity of V-PBFT is O ( N ) . • It improves robustness by combining the view-change process with credit mechanism. • Followers prioritize to ask correct client requests instead of lunching view changes. • Its consensus performance is better than that of the classic HotStuff and PBFT.
Authors
- Anqi Li (ORCID: https://orcid.org/0000-0003-1864-8035)
- Q. Y. Zhang (ORCID: https://orcid.org/0009-0003-5625-3779)
- Yingbiao Yao (ORCID: https://orcid.org/0000-0001-7946-6070)
- Xin Xu
Institutions
- Hangzhou Dianzi University (CN)
Publication Details
- Journal
- Computer Networks
- Published
- 2025-04-25
- DOI
- https://doi.org/10.1016/j.comnet.2025.111307
- Primary Topic
- Complex Systems and Time Series Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00