The network management is one of the most expensive tasks for telecommunications operators. Consequently, there is a tendency to delegate the management of the network in the network itself. This is known as Autonomic Networking. But today, fault diagnosis remains a non-autonomous task. Traditionally this has been done by human experts supported by monitoring systems for detecting alarms or symptoms. But even with these systems, the diagnostic task is primarily a manual process.
The constant increase in size and complexity of the network makes fault diagnosis is a critical task for the business that must be managed quickly and reliably. To carry out highly qualified engineers are required, but even these people are not always able to cope with the increasing diversity and complexity of networks, since the diagnosis is a difficult, time consuming and, therefore, it is a costly task. Consequently, operators aim to fully automate fault diagnosis to reduce operating costs and improve the customer experience through the automated operation of standardized diagnostic processes.
In order to overcome the above limitations and advance the development of techniques for automatic diagnosis for SDN networks, this project aims to generate different monitoring techniques and diagnostics based on artificial intelligence techniques that allow the system to learn or adapt to changes and developments in the network. To do they arise study and evaluate different techniques of evolutionary computation and machine learning techniques.
On the one hand, evolutionary computing techniques are very attractive and techniques used in the literature to solve different optimization problems, which is interesting to evaluate and / or monitor the status of the network and to detect different anomalies in real time. On the other hand, machine learning techniques offer the ability to learn from the environment system with previously collected data, which allows to adapt the system behavior if the environment is modified by external factors.
Project TSI-100102-2016-12 project funded by the Ministry of Industry, Energy and Tourism.