基于边缘计算的收益激励算法对区块链分片的优化

Optimization of blockchain sharding by profit incentive algorithm based on edge computing

本文提出一种收益激励(PI)算法,首先针对单个节点计算其生成一个区块的延迟和能耗,计算出该节点处理任务的最终收益;其次根据基于边缘计算的可信度模型计算出该节点的平均分片可信度,在不降低其他节点收益和平均分片可信度的同时,该节点选择能够最大化其收益的分片;最后得到所有节点收益和分片可信度均最大化的分片结构。仿真结果表明,在基于边缘计算的区块链分片中,与OmniLedger、DBSS和CBBA算法进行比较,PI算法能够在保证吞吐量不降低的情况下,提高区块链系统的稳定性。

A Profit Incentive(PI)algorithm is proposed.First,the delay and energy consumption of a block generated by a single node are calculated,and the final profit of the node is calculated according to the delay and energy consumption.Next,the average partition credibility of the node is calculated according to the credibility model based on edge computing.Next,the node chooses the partition that can maximize its profit without reducing the other nodes'profit and average partition credibility.Finally,a stable sharding structure is obtained to maximize the profit of all nodes and the reliability of sharding.Simulation results show that,compared with OmniLedger,domain based sharding scheme and credibility based blockchain algorithm,PI algorithm can improve the stability of blockchain sharding without reducing throughput.