Computer Science > Networking and Internet Architecture
[Submitted on 5 Jun 2013 (this version), latest version 15 Jan 2015 (v3)]
Title:CSMA using Statistical Physics toward Throughput and Utility Optimal CSMA
View PDFAbstract:In the recent past years, CSMA (Carrier Sense Multiple Access), which resolves contentions over wireless networks in a fully distributed fashion, has gained a lot of attentions since it has been proved in several papers that appropriate control of CSMA parameters provably guarantees optimality in terms of throughput and fairness. Most algorithms rely on the famous MCMC (Markov Chain Monte Carlo) technique, which enables one to find optimal CSMA parameters through iterative loops of simulation-and-update. However, such a simulation-based approach often becomes a major cause of slow convergence, being vulnerable to flow/topology changes. In this paper, we develop distributed iterative algorithms which produce approximate solutions with fast convergence. Our motivating tools are variational methods, referred to as Bethe and Kikuchi approximations in statistical physics, allowing us to express approximate solutions via a small number of certain non-linear equations. Fast convergence comes from directly solving those non-linear equations in a distributed manner, rather than multiple loops of simulation-and-update in existing algorithms. Our simulation results verify our theoretical developments and show that our approximate solutions are highly accurate and converge much faster than prior algorithms under various scenarios.
Submission history
From: Seyoung Yun [view email][v1] Wed, 5 Jun 2013 12:17:41 UTC (420 KB)
[v2] Fri, 22 Nov 2013 04:51:41 UTC (337 KB)
[v3] Thu, 15 Jan 2015 15:14:04 UTC (425 KB)
Current browse context:
cs.NI
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.