DPoS共识机制的信誉优化OA北大核心

A reputation evaluation consensus mechanism RE-DPoS is proposed to address the passive node voting,node collusion,and high probability of malicious nodes being selected as agent nodes in the DPoS consensus mechanism.First,the voting mechanism of nodes is optimized by introducing verifiable random functions to vote on them,limiting their negative voting behavior.Second,the evaluation method of nodes is improved by dynamically calculating the weights of various indicators of nodes through entropy weight method,which more accurately evaluates the reputation status of nodes and ensures that the selected agent nodes have higher reputation.Finally,a reward and punishment mechanism is added to update the reputation value based on node behavior,and a binary exponential backoff algorithm is introduced to set the number of waiting rounds for malicious nodes to encourage them to actively participate in voting and effectively handle malicious nodes.The experimental results show that RE-DPoS effectively reduces the probability of malicious nodes being elected as agent nodes,improves node voting,and thus ensures the security of blockchain systems.