DocumentCode
2403853
Title
Applying neural networks and other AI techniques to fault detection in satellite communication systems
Author
Elerin, Liese ; Learoyd, Charles ; Wilson, Beth
Author_Institution
Raytheon Co., Marlborough, MA, USA
fYear
1997
fDate
24-26 Sep 1997
Firstpage
617
Lastpage
625
Abstract
A demonstration program has been completed to apply various artificial intelligence techniques including, neural networks, expert systems, and case-based reasoning to fault detection in satellite communications systems. The GMM program implemented these techniques for global military satellite communications maintenance. Neural networks were designed and trained to analyze incoming built-in-test (BIT) fault signatures from the satellite communications terminal. Expert systems were developed to embed diagnostic knowledge relating to equipment maintenance. The prototype hybrid system uses neural filters to detect faults, which are further processed by expert systems to classify the faults and provide repair directions
Keywords
case-based reasoning; diagnostic expert systems; fault diagnosis; maintenance engineering; military communication; neural nets; satellite communication; AI techniques; GMM program; built-in-test fault signatures; case-based reasoning; diagnostic knowledge; equipment maintenance; expert systems; fault detection; global military satellite communications maintenance; neural filters; neural networks; repair directions; satellite communication systems; Artificial intelligence; Artificial neural networks; Costs; Diagnostic expert systems; Fault detection; Filters; History; Intelligent networks; Neural networks; Satellite communication;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks for Signal Processing [1997] VII. Proceedings of the 1997 IEEE Workshop
Conference_Location
Amelia Island, FL
ISSN
1089-3555
Print_ISBN
0-7803-4256-9
Type
conf
DOI
10.1109/NNSP.1997.622444
Filename
622444
Link To Document