Journals
  Publication Years
  Keywords
Search within results Open Search
Please wait a minute...
For Selected: Toggle Thumbnails
d-PBFT:detection consensus algorithm for alliance blockchain
LIU Yu, ZHU Chaoyang, LI Jinze, LAO Yuanji, QIN Tuanfa
Journal of Computer Applications    2021, 41 (3): 756-762.   DOI: 10.11772/j.issn.1001-9081.2020060900
Abstract364)      PDF (1007KB)(1322)       Save
There is an identity authentication mechanism in alliance blockchain, but even by using the mechanism, Byzantine malicious nodes still exist in the network, and the member nodes in the alliance may be controlled and utilized by the third-party enemies. To solve these problems, a detection-Practical Byzantine Fault Tolerance (d-PBFT) consensus algorithm that can monitor the node states was proposed. Firstly, the primary node was elected and the its state was checked to ensure that the elected primary node never be malicious before. Secondly, the consensus request submitted by the client was executed through the three-stage consensus process "pre-prepare-prepare-commit". Finally, the primary node state was assessed according to the three-stage achievement state. If primary node was unstable or malicious, it would be marked, and the malicious node would be added to the quarantine to wait for processing. In this algorithm, based on tolerating a specific number of Byzantine nodes, every node state was monitored all the time and the malicious nods would be isolated. In this case, the bad impact of malicious nodes on the alliance would be reduced. Experimental results show that the network with d-PBFT algorithm has high throughput and low consensus delay, and the consensus generation amount of the algorithm is 26.1% more than that of Practical Byzantine Fault Tolerance (PBFT) algorithm when alliance network includes Byzantine nodes. The d-PBFT algorithm not only improves the robustness of alliance network, but also improves the network throughput.
Reference | Related Articles | Metrics