Centdian computation for sensor networks

Boaz Ben-Moshe*, Amit Dvir, Michael Segal, Arie Tamir

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

This paper focuses on the centdian problem in a cactus network where a cactus network is a connected undirected graph, and any two simple cycles in the graph have at most one node in common. The cactus network has important applications for wireless sensor networks when a tree topology might not be applicable and for extensions to the ring architecture. The centdian criterion represents a convex combination of two QoS requirements: transport and delay. To the best of our knowledge, no efficient algorithm has yet been developed for constructing a centdian node in a cactus graph, either sequential or distributed. We first investigate the properties of the centdian node in a cycle graph, and then explore the behavior of the centdian node in a cactus graph. Finally, we present new efficient sequential and distributed algorithms for finding all centdian nodes in a cycle graph and a cactus graph.

Original languageEnglish
Title of host publicationTheory and Applications of Models of Computation - 7th Annual Conference, TAMC 2010, Proceedings
Pages187-198
Number of pages12
DOIs
StatePublished - 2010
Event7th Annual Conference on Theory and Applications of Models of Computation, TAMC 2010 - Prague, Czech Republic
Duration: 7 Jun 201011 Jun 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6108 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference7th Annual Conference on Theory and Applications of Models of Computation, TAMC 2010
Country/TerritoryCzech Republic
CityPrague
Period7/06/1011/06/10

Keywords

  • Cactus Graph
  • Centdian Node
  • Distributed Algorithm
  • Sensor Network

Fingerprint

Dive into the research topics of 'Centdian computation for sensor networks'. Together they form a unique fingerprint.

Cite this