Repository logo
Log In(current)
  1. Home
  2. Colleges & Schools
  3. Graduate School
  4. Doctoral Dissertations
  5. A knowledge-based system for design of broadband local area networks
Details

A knowledge-based system for design of broadband local area networks

Date Issued
August 1, 1989
Author(s)
Yi, Kyŏng-su
Advisor(s)
William G. Sullivan
Additional Advisor(s)
Asa O. Bishop
Elden L. DePorter
William T. Snyder
Osama Soliman
Robert E. Uhrig
Permanent URI
https://trace.tennessee.edu/handle/20.500.14382/19961
Abstract

The objective of this research was to build a prototype knowledge-based (KB) system for the design of a broadband cable local area network (LAN) in accordance with the MAP/TOP Broadband Specification for computer integrated manufacturing.


An overall broadband cable network design is a complex task because many conflicting variables and objectives must be satisfied. The approach taken in this research was to divide a network design problem into three major hierarchical steps. First, the topology of a network was determined. Second, based on the designed topology of the network, system parameters were determined. Third, based on the previous results, active and passive devices were laid out along the cable route.

Because of the uniqueness of a broadband LAN design, it is difficult to apply existing KB systems to broadband LAN design. One of the difficulties in applying KB systems to design concerns how best to represent design problems. The design of a broadband network is an open-ended problem in which it is not known in advance how many nodes there will be in the network. The method developed in this research was to decompose the tree structure of a network into a series of parent-child relationships. Without this general representation scheme, it is difficult to apply a KB system to the network design problem because there are an infinite number of tree networks.

Another problem of applying an existing KB system to the system layout of broadband cable network design is made difficult by the fact that an inference engine must be applied repeatedly to each link of a network. A control scheme was developed for this task.

The contribution made by this dissertation is establishing the feasibility of a KB system for broadband cable network design and automating parts of broadband design with knowledge and heuristics by actually developing a prototype system.

Degree
Doctor of Philosophy
Major
Engineering Science
File(s)
Thumbnail Image
Name

Thesis89b.L332.pdf

Size

9.63 MB

Format

Unknown

Checksum (MD5)

c7b5d10bb9cee10bfa569515028f3237


University Libraries

1015 Volunteer Boulevard
Knoxville, TN 37996
865-974-4351

Map & Directions
Donate to the Libraries
  • About
  • John C. Hodges Society
  • Speaking Volumes magazine
  • Outreach
  • Directory
  • Employment
  • Policies
  • Library Intranet
University of Tennessee power T logo

The University of Tennessee, Knoxville
Knoxville, Tennessee 37996
865-974-1000

Events
A-Z
Apply
Privacy
Map
Directory
Give to UT
Accessibility

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science