Communication Network Failure Detection Utilizing
Artificial Neural Networks

C.B. Hudgins, B.C. Drake, S.K. Biswas and H.M. Sendaula


Abstract
********

There is increasing dependence on communication networks in the
day-to-day operation in all types of organizations.  Thus, it has
become increasingly important for these communication networks
to be reliable systems.  Due to the complexity of these networks,
human network managers can no longer ensure network reliability
without electronic tools to help them detect or predict faults.  In
this paper, a fault detection system based on an artificial neural
network is presented for prompt detection of faults in communication
networks.  A simulation of this system based on a multi-layer
feed-forward neural network is developed.  The fault detection system
is trained from data generated from a queuing model of a four node
network.  Once trained, the fault detection system is able detect
faults in the network model.