Research Article Open Access

Fault Detection of Computer Communication Networks Using an Expert System

Ibrahiem M.M. El Emary and Adanan I. Al Rabia

Abstract

The main objective of this study was to build an expert system for assisting the network administrator in his work of management and administration of the computer communication network. Theory of operation of the proposed expert system depends on using a time series model capable of forecasting the various performance parameters as: delay, utilization and collision frequency. When the expert system finds a difference (with certain tolerance) between the predicted value and the measured value, it informs the network administrator that there exist problems in his network either in the switch or link or router. We examine two types of network by our proposed expert system, the first one is called token bus while the second one is called token ring. When we run our expert system on these two types of computer networks, actually the expert system captures the problem when there exists an excess deviation from the network performance parameters.

American Journal of Applied Sciences
Volume 2 No. 10, 2005, 1407-1411

DOI: https://doi.org/10.3844/ajassp.2005.1407.1411

Submitted On: 24 October 2005 Published On: 31 October 2005

How to Cite: El Emary, I. M. & Al Rabia, A. I. (2005). Fault Detection of Computer Communication Networks Using an Expert System. American Journal of Applied Sciences, 2(10), 1407-1411. https://doi.org/10.3844/ajassp.2005.1407.1411

  • 3,624 Views
  • 2,802 Downloads
  • 3 Citations

Download

Keywords

  • Knowledge base
  • inference engine
  • exponential weighted moving average
  • user interface
  • forward chaining