Data-oriented Networks

The current network ecosystem is passing through a very large paradigm shift from protocol enhancements towards network behavior optimization. This is mainly due to the reaching of the maximum optimization potential of the ubiquitous uniform network model. To be able to move forward, resources will have to be scheduled to be able to address the diverse requirements.

This can be achieved only by having a holistic perspective of the system, knowing the impact of any schedule on the rest of the system. Also, this type of optimization for the communication systems is highly specific. To be able to trigger such operations, first, there is a need for a large amount of local data acquisition and its processing. This is achieved with the new advanced monitoring and data distribution technologies. Using this data, new algorithms must be defined which account for the specificity and at the same time build new handling processes. 

Data-oriented Networks Graph
Fraunhofer FOKUS

For being able to automate the acquisitions, storage, curation, and exchange of the data between the different active systems, network management, and insight generating elements (e.g. machine learning algorithm) careful attention has to be given towards the development of a comprehensive data layer. For more information on this research direction, please check the NEMI-Website.