中国物理B ›› 2010, Vol. 19 ›› Issue (12): 120513-120513.doi: 10.1088/1674-1056/19/12/120513

• • 上一篇    下一篇

Convergence speed of consensus problems over undirected scale-free networks

孙巍, 窦丽华   

  1. School of Automation, Beijing Institute of Technology, Beijing 100081, China
  • 收稿日期:2010-03-09 修回日期:2010-08-08 出版日期:2010-12-15 发布日期:2010-12-15
  • 基金资助:
    Project supported by the National Natural Science Foundation for Distinguished Young Scholars of China (Grant No. 60925011).

Convergence speed of consensus problems over undirected scale-free networks

Sun Wei(孙巍)†ger and Dou Li-Hua(窦丽华)   

  1. School of Automation, Beijing Institute of Technology, Beijing 100081, China
  • Received:2010-03-09 Revised:2010-08-08 Online:2010-12-15 Published:2010-12-15
  • Supported by:
    Project supported by the National Natural Science Foundation for Distinguished Young Scholars of China (Grant No. 60925011).

摘要: Scale-free networks and consensus behaviour among multiple agents have both attracted much attention. To investigate the consensus speed over scale-free networks is the major topic of the present work. A novel method is developed to construct scale-free networks due to their remarkable power-law degree distributions, while preserving the diversity of network topologies. The time cost or iterations for networks to reach a certain level of consensus is discussed, considering the influence from power-law parameters. They are both demonstrated to be reversed power-law functions of the algebraic connectivity, which is viewed as a measurement on convergence speed of the consensus behaviour. The attempts of tuning power-law parameters may speed up the consensus procedure, but it could also make the network less robust over time delay at the same time. Large scale of simulations are supportive to the conclusions.

Abstract: Scale-free networks and consensus behaviour among multiple agents have both attracted much attention. To investigate the consensus speed over scale-free networks is the major topic of the present work. A novel method is developed to construct scale-free networks due to their remarkable power-law degree distributions, while preserving the diversity of network topologies. The time cost or iterations for networks to reach a certain level of consensus is discussed, considering the influence from power-law parameters. They are both demonstrated to be reversed power-law functions of the algebraic connectivity, which is viewed as a measurement on convergence speed of the consensus behaviour. The attempts of tuning power-law parameters may speed up the consensus procedure, but it could also make the network less robust over time delay at the same time. Large scale of simulations are supportive to the conclusions.

Key words: scale-free networks, consensus, power-law distribution

中图分类号:  (General topology)

  • 02.40.Pc
89.75.Hc (Networks and genealogical trees)