Research On Blockchain Consensus Algorithm For Internet Of Things

Due to the characteristics of decentralization and non-tamperable data,blockchain has been widely used in various fields,and the Internet of Things is one of them.However,the current blockchain system cannot meet the performance requirements of Io T application scenarios in terms of time delay and so on.For a blockchain system,performance such as delay depends largely on the performance of the consensus algorithm used.At present,the main consensus algorithms are Po W,Po S,DPo S,Paxos,Raft,PBFT,etc.The consensus algorithm represented by Po W and Po S has the disadvantages of high energy consumption,low efficiency,and dependence on tokens.Traditional consensus algorithms,such as Paxos and Raft,do not consider Byzantine fault tolerance.However,the PBFT algorithm has the problems of random selection of the master node and taking up too much network resources.After analyzing and studying the above consensus algorithm,the PBFT algorithm is improved.The improvements are as follows:1.In view of the small number of nodes,a PBFT algorithm based on the ranking and ranking mechanism is proposed.Based on the original PBFT algorithm,the algorithm adds a scoring ranking mechanism and improves the master node selection algorithm.After each round of consensus is completed,the nodes are rewarded or punished according to their behavior.The probability that a node is selected as the primary node is related to the ranking of the node's total score.Experimental results show that the improved algorithm reduces the probability of malicious nodes being selected as the master node and increases the probability of high-quality nodes being selected as the master node,while reducing the consensus delay.2.In view of the large number of nodes,a PBFT algorithm based on partition clustering is proposed.The algorithm combines the clustering algorithm based on the PBFT algorithm based on the scoring sorting mechanism,divides all nodes into several clusters based on the node routing hops and the network delay between the nodes,and then uses the PBFT algorithm based on the scoring sorting mechanism for consensus.The local consensus of the cluster proceeds simultaneously.Experimental results show that the improved algorithm greatly reduces the communication volume and consensus delay.