--------------------------------------------------------------- 
|                                                               |
|                  Stephen W. Neville                           |
|          A Prototype Expert System Based Fault                |
|      Diagnosis System for Cable Trunk Amplifier Plants        |
|                                                               |
|                         MASc THESIS                           |
|      Department of Electrical and Computer Engineering        |
|                  University of Victoria                       |
|                          1992                            |
|                                                               |
 --------------------------------------------------------------- 


	As the complexity of communication networks 
increases, it becomes increasingly difficult to isolate 
and diagnose network faults. This maintenance problem is
 due to the large number of functional elements within 
the networks, the elements' technical complexity, and 
their inherent interaction. One technique to solve this 
problem is to use expert system technology to manage 
the vast amount of status information produced by these 
networks and to provide an automated diagnostic reasoning 
about the location and cause of observed faults. In this 
thesis, a prototype expert system based diagnostic tool 
for cable trunk amplifier networks is presented. This 
tool is capable of isolating and diagnosing the most 
common network level faults which occur in these type 
of networks. This tool was created as a proof-of-principle 
of the use of expert system based diagnostic tools on 
this type of communication network. As such, the tool
 was tested on a series of network faults occurring 
within the Rogers Cablesystems Ltd. Victoria cable trunk 
amplifier network. The results of these tests are presented.