UNificatins and Extensions of the Multiple Access Communications Problem,

Abstract

Multiple access protocols permit a broadcast communications channel to be shared by a large number of stations under distributed control. It is assumed that only one message at a time can be transmitted successfully over the common channel. The author derives a local optimality condition for synchronous multiple access protocols, and show that many known protocols are special cases of this condition. Included a survey of much of the recent work on infinite population tree algorithms that use the history of channel activity to carry out short-range dynamic scheduling. A novel approach is presented for deriving upper bounds on the maximum stable throughput with finite average delay for infinite population protocols. Bounds are found for the case of arbitrarily complex algorithms, and for the restricted (but reasonable) class of protocols that obey a degenerate intersection property. This latter class is quite interesting, being a slight generalization of first-come first-served that includes all currently proposed protocols. The model is extended to include multiple access protocols aided by a partial reservation channel. Particular emphasis is placed on multiple access in the context of local networks.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1981
Accession Number
ADA130785

Entities

People

  • Mart Lauri Molle

Organizations

  • University of California, Los Angeles

Tags

Communities of Interest

  • Energy and Power Technologies
  • Materials and Manufacturing Processes
  • Space

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Satellites
  • Coaxial Cables
  • Communication Systems
  • Computer Communications
  • Computer Networks
  • Computer Programming
  • Computer Science
  • Computers
  • Engineering
  • Information Processing
  • Local Area Networks
  • Multiple Access
  • Network Science
  • Optical Fibers
  • Packet Switching
  • Random Variables

Fields of Study

  • Computer science
  • Mathematics

Readers

  • Computer Networking
  • Mathematical Modeling and Probability Theory.