A Bayesian-based Self-Diagnosis Approach for Alarm Prognosis in Communication Networks

Abstract

Alarms are the symptom of faults and indicate an abnormal of the communication networks. The Bayesian network is one of the most powerful and popular fault analysis tools. In this paper, we present the alarm prognosis method based on Bayesian inference in order to estimate the health state and trend of the network given by an alarm take as evidence enters the network. The approach reduces human intervention and enhances the availability of the work effectively. Finally, experimental results verify the validity of the approach.