Open AccessComputing weighted value of the fast linear iterations distributed consensus algorithm based on label propagation algorithm
Author(s) -
Peng Huanxin,
Bin Liu
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/631/5/052005
Subject(s) - algorithm , convergence (economics) , distributed algorithm , rate of convergence , computer science , network topology , node (physics) , consensus , mathematics , topology (electrical circuits) , multi agent system , distributed computing , artificial intelligence , key (lock) , computer security , structural engineering , combinatorics , engineering , economics , economic growth , operating system
In order to accelerate the convergence rate of distributed consensus under complex topology, the fast linear iterations distributed consensus algorithm was proposed by Xiao L. the fast linear iterations distributed consensus algorithm improved dramatically the convergence rate of the distributed consensus problem, but computing the weighted value was difficult, and the computing volume geometrically increased with the expansion of the communication topology. In the paper, the weighted value of every node was computed based on label propagation algorithm. In the paper, firstly, the complex network was divided into a few communities by the label propagation algorithm, the nodes between different community were weighted. The analysis and simulation of computing volume and the convergence speed were done. The computing volume of the proposed method was smaller than that of the fast linear iterations distributed consensus algorithm.