Publication detail

Influence of random leader appointment on convergence rate of network size estimation

KENYERES, M. KENYERES, J.

Original Title

Influence of random leader appointment on convergence rate of network size estimation

English Title

Influence of random leader appointment on convergence rate of network size estimation

Type

journal article in Web of Science

Language

en

Original Abstract

The information about the network size is crucial for many real life applications. It can be obtained with the distributed average consensus algorithm, whose implementation requires the proper leader appointment, which is an energy demanding process. The lack of the papers concerned with this aspect motivates us to verify the influence of a random leader appointment on the convergence rates of different weight models of average consensus. We examine the range of the achieved convergence rates in 30 randomly generated networks for different leaders and show the maximal possible deceleration of the algorithm due to an inappropriate leader appointment.

English abstract

The information about the network size is crucial for many real life applications. It can be obtained with the distributed average consensus algorithm, whose implementation requires the proper leader appointment, which is an energy demanding process. The lack of the papers concerned with this aspect motivates us to verify the influence of a random leader appointment on the convergence rates of different weight models of average consensus. We examine the range of the achieved convergence rates in 30 randomly generated networks for different leaders and show the maximal possible deceleration of the algorithm due to an inappropriate leader appointment.

Keywords

Distributed computing, average consensus algorithm, network size estimation, leader appointment

Released

01.12.2017

Pages from

57

Pages to

68

Pages count

12

URL

BibTex


@article{BUT141242,
  author="Martin {Kenyeres} and Jozef {Kenyeres}",
  title="Influence of random leader appointment on convergence rate of network size estimation",
  annote="The information about the network size is crucial for many real life applications. It can be obtained with the distributed average consensus algorithm, whose implementation requires the proper leader appointment, which is an energy demanding process. The lack of the papers concerned with this aspect motivates us to verify the influence of a random leader appointment on the convergence rates of different weight models of average consensus. We examine the range of the achieved convergence rates in 30 randomly generated networks for different leaders and show the maximal possible deceleration of the algorithm due to an inappropriate leader appointment.",
  chapter="141242",
  howpublished="online",
  number="4",
  volume="79",
  year="2017",
  month="december",
  pages="57--68",
  type="journal article in Web of Science"
}