当前位置: X-MOL 学术Big Data Res. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Proof-of-Activity Consensus Algorithm Based on K-Medoids Clustering
Big Data Research ( IF 3.3 ) Pub Date : 2021-08-30 , DOI: 10.1016/j.bdr.2021.100266
Dong Wang 1, 2 , Chenguang Jin 1 , Bingbing Xiao 1 , Zheng Li 1 , Xin He 1, 2
Affiliation  

K-medoids cluster-based Proof of Action consensus algorithm (KPoA) is proposed by us in order to mainly decrease the computing power wast and Malicious information dissemination in Proof of Activity consensus algorithm (PoA). In PoA, offline participant nodes cause block headers to be discarded, then the terrible wastage of computing power takes place. Moreover, the efficiency of consensus is greatly affected because the malicious nodes are not handled in time. KPoA uses K-medoids clustering and follow-the-satoshi mechanism to select participant nodes and accounting nodes successively to ensure the unpredictability of accounting nodes. All nodes are grouped by K-medoids clustering, which help to achieve transaction blocks' hierarchical consensus and reduce the spread of malicious information. Multiple accounting nodes take turns to create transaction blocks in the same block header, which nicely increases the number of transaction blocks and reduces the waste of computing power. In order to improve the activity of nodes, the reward and punishment scheme based on credit is set up. The credit value is used to adjust the probability of a node creating blocks. Also malicious nodes can be eliminated in time according to credit value. Experimental results show that KPoA creates transaction blocks at an acceleration of about 2.5 times that of PoA. And its stability is better than PoA.



中文翻译:

基于 K-Medoids 聚类的活动证明共识算法

我们提出了基于K-medoids集群的行动证明共识算法(KPoA),主要是为了减少活动证明共识算法(PoA)中的计算能力浪费和恶意信息传播。在 PoA 中,离线参与者节点会导致区块头被丢弃,从而发生可怕的算力浪费。而且,由于恶意节点没有得到及时处理,极大地影响了共识的效率。KPoA采用K-medoids聚类和follow-the-satoshi机制依次选择参与者节点和记账节点,保证记账节点的不可预测性。所有节点通过K-medoids聚类进行分组,有助于实现交易区块的分层共识,减少恶意信息的传播。多个记账节点轮流在同一个区块头中创建交易区块,很好的增加了交易区块的数量,减少了算力的浪费。为了提高节点的活跃度,建立了基于信用的奖惩方案。信用值用于调整节点创建区块的概率。也可以根据信用值及时消除恶意节点。实验结果表明,KPoA 创建交易块的速度大约是 PoA 的 2.5 倍。并且其稳定性优于PoA。信用值用于调整节点创建区块的概率。也可以根据信用值及时消除恶意节点。实验结果表明,KPoA 创建交易块的速度大约是 PoA 的 2.5 倍。并且其稳定性优于PoA。信用值用于调整节点创建区块的概率。也可以根据信用值及时消除恶意节点。实验结果表明,KPoA 创建交易块的速度大约是 PoA 的 2.5 倍。并且其稳定性优于PoA。

更新日期:2021-09-08
down
wechat
bug