Fair and flexible budget-based clustering

  • Authors:
  • Fehmi Ben Abdesslem;Artur Ziviani;Marcelo Dias De Amorim;Petia Todorova

  • Affiliations:
  • UPMC, Paris, France;National Laboratory for Scientific Computing, Petrópolis, Brazil;UPMC, Paris, France;Fraunhofer FOKUS, Berlin, Germany

  • Venue:
  • ICC'09 Proceedings of the 2009 IEEE international conference on Communications
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

An efficient way to bound the size of clusters in large-scale self-organizing wireless networks is to rely on a budget-based strategy. The side effect of conventional budget-based clustering approaches is that they generate a potentially large number of small, even single-node, clusters. The consequence is that while clusters are bounded, their average size may be far from the expected value (the budget), which negatively impacts the performance of the communication systems running on top of it. In contrast, we propose Fair and Flexible Budget-Based Clustering (FFBC) to form size-controlled clusters in large-scale self-organizing networks. For a given target cluster size, our approach outperforms previous budget-based algorithms by creating clusters of average size closer to the requested value and avoiding isolated nodes.