Carrier-Sense Stack Algorithms for Multiple Access Communication Channels.

Abstract

We consider the random multiple access of a collision-type, packet-switched channel, for the Poisson user model in a local area network environment, where 'carrier sensing' techniques are possible due to small propagation delays. We propose and analyze random access algorithms that are representative of a new class of stable algorithms with 'limited sensing' and 'free access' characteristics. 'Limited sensing' algorithms require that users sense the channel only while they have a packet to transmit, and, therefore, they have practical advantages over algorithms that require continuous channel sensing. The 'free access' characteristics of the proposed algorithms simplify their implementation, since newly arrived packets are transmitted upon arrival, provided that the channel is sensed idle. Utilizing the regenerative character of the stochastic processes that are associated with the random access system, we derive lower bounds on the maximum stable throughput, and tight upper and lower bounds on the induced mean packet delay. The proposed algorithms are easy-to-implement, and they combine inherently stable operation and high performance with modest channel sensing requirements.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1986
Accession Number
ADA169774

Entities

People

  • Lazaros Merakos

Organizations

  • University of Connecticut

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Boundaries
  • Cables
  • Channel Models
  • Coefficients
  • Collisions
  • Communication Channels
  • Computer Science
  • Electrical Engineering
  • Engineering
  • Local Area Networks
  • Multiple Access
  • Networks
  • Probability
  • Random Variables
  • Stochastic Processes
  • Transmitting

Fields of Study

  • Computer science

Readers

  • Radio communications and signal processing.
  • Wave Propagation and Nonlinear Chaotic Dynamics.