Research on Immune Agents for Network Fault Diagnosis

This paper proposes a Network Fault Diagnosis Model based on combination of immune algorithm and agent technology, which analysis of biological immunity mechanism in-depth and study of agent technology with highly adaptive, self-control, self-learning for the above characteristics of network and its failure presently. In a distributed environment, in accordance with the different characteristics of the network and different environmental, the agents with different functions and categories diagnose the faults. Multi-agents collaborate among each other, exchange information, and collect data, in order to diagnose the types of fault. Experiments show that the method can determine the types of network fault effectively, which is a good reference for fault diagnosis of large-scale networks in the future.