Employing Recurrent Artificial Neural Networks for Developing Baselines for Proactive Network Management

摘要

This work presents a methodology to develop autonomous agents for network management. There are two kinds of agents to develop: static or dynamic agents. The first one can be implemented; using heuristics obtained from an expert or the network administrator, through production rules or feed forward neural networks (NN). Using the network examples we can construct dynamic agents. The NN may be trained to solve a problem using some examples. Moreover, the behavior of the management must be considered, the network management may be reactive or proactive. Normally, we have the reactive behavior when the problem occurs and after we will search for a solution. We may see in diagnostic or troubleticket systems for Fault Management. On the contrary, the proactive behavior is a preventive control of the network. We divided the network management in the five functional areas proposed by OSI Model Reference. Thus, each area has a different intelligent solution.