Distributed Fault Detection and Localization Algorithm Using Artificial Neural Network in Optical WDM Networks
Fault identification and localization is an important network management function in an optical WDM network which has an impact on survivability of the network. And so, it is critical for optical networks to employ fast and effective methods for detecting and locating network failures. In this paper, a distributed agent based fault management algorithm is proposed. The distributed agents are based on Artificial Neural Networks (ANN) which detect the failures and guarantee proactive fast re-routing. In ANN, the input set is represented by string of four fields which contain the node ID, type of component and position of the component inside the node.