Time efficient gossiping in known radio networks

Leszek Ga̧sieniec, Igor Potapov, Qin Xin

Research output: Chapter in Book/Report/Conference proceedingChapter

27 Scopus citations

Abstract

We study here the gossiping problem (all-to-all communication) in known radio networks, i.e., when all nodes are aware of the network topology. We start our presentation with a deterministic algorithm for the gossiping problem that works in at most n units of time in any radio network of size n. This is an optimal algorithm in the sense that there exist radio network topologies, such as: a line, a star and a complete graph in which the radio gossiping cannot be completed in less then n units of time. Furthermore, we show that there isn't any radio network topology in which the gossiping task can be solved in time < ⌊log(n - 1)⌋ + 2. We show also that this lower bound can be matched from above for a fraction of all possible integer values of n; and for all other values of n we propose a solution admitting gossiping in time ⌈log(n - 1)⌉ + 2. Finally we study asymptotically optimal O(D)-time gossiping (where D is a diameter of the network) in graphs with max-degree Δ = O(D1-1/(i+1)/ logi n), for any integer constant i ≥ 0 and D large enough.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
EditorsRastislav Kralovic, Ondrej Sykora
PublisherSpringer Verlag
Pages173-184
Number of pages12
ISBN (Print)3540222308
DOIs
StatePublished - 2004
Externally publishedYes

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3104
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Time efficient gossiping in known radio networks'. Together they form a unique fingerprint.

Cite this