伪多跳中继分布式一致性算法

为了提高分布式一致性问题的收敛速度, 减少节点间的通信成本, 本文提出一种分布式一致性算法. 新算法采用单跳通信, 利用非邻接节点的前状态信息进行节点状态更新. 本文证明了在无向通信拓扑下新算法的一致性收敛, 并分析计算了新算法的通信量以及收敛速度的变化. 通过理论分析和仿真验证, 结果表明新算法具有收敛速度快, 通信方式简单, 通信量少, 存在通信延时情况下通信数据不容易丢失等优点.

Pseudo multi-hop relay distributed consensus algorithm

In order to accelerate the convergence rate of distributed consensus problems and reduce the communication costs among nodes. A new distributed consensus algorithm with single-hop communication is presented. The new algorithm makes use of the previous information of non-adjacent nodes to update the states of the nodes. We prove that the algorithm is convergent under undirected topology, and calculate the communication volume and convergence rate of the new algorithm. By analyzing and simulating on the new algorithm, the convergence rate of the new algorithm is higher, the communication mode is simpler and the communication overload is less in the new algorithm, and communication data are not easy to lose under communication time-delays.