Bayesian Networks Based Alarm Correlation and Fault Diagnosis in Communication Networks
基于贝叶斯网络的通信网告警相关性和故障诊断模型

This paper proposes the alarm correlation and fault identification based on Bayesian networks in communication networks. At first, the basic concepts of Bayesian networks are introduced. Then the paper presents an approach for modeling large communication networks that are divided into their constituting sub-networks. And the causal relation is used to model the functional relationship among the sub-networks. The paper discusses how to construct Bayesian networks from the causal relation and presents a distributed alarm correlation framework based on CORBA. Finally, the realization and results of alarm correlation and fault identification is discussed in SDH over DWDM systems. The experimentation has proved that using Bayesian network based alarm correlation is benefit to detect and localize the root faults in communication networks.