Collaborative Research Centre 1053

Multi-Mechanisms Adaption for the Future Internet

The Collaborative Research Centre (CRC) Multi-Mechanisms Adaption for the Future Internet (MAKI) addresses questions about the future of the internet and its implications for communication patterns and demands on the underlying infrastructure. Since 2013, CRC MAKI concerns itself with the mechanisms in communication systems and constant development thereof.

Project area B assesses the state and quality of a communication systems as analysis of this data allows further planning of transitions. Katharina Keller is involved in Subproject B3, which examines transitions in communication systems from an economic cross-layer perspective and provides decentralized analysis and planning solutions for autonomous end devices. It allows an efficient and realistic application of proactive transitions for coexisting multi-mechanisms wireless networks with software-defined autonomous nodes by acknowledging user preferences.

Project area C conducts research on communication mechanisms applicable to transition-capable communication systems. Subproject C5, with the contribution of Patrick Weber, analyzes anticipatory networks that can proactively optimize their operation based on predicted events by making use of sensory functions of mobile devices and availability of relevant data on the Internet. Novel methods to classify and predict exogenous events, that bring about a large chance in number of network participants, will be developed. Research will be conducted on extraction of rules to detect such events and their execution on an probabilistic, proactive and mobile CEP engine located on wireless nodes at the edge of the Internet to develop proactive transitions.

In the transfer projects, basic scientific results from the CRC MAKI will be transferred to various application areas together with industrial partners. The transfer projects are expected to have repercussions on issues in basic research in the CRC. The transfer project T2, with the contribution of Dr. Patrick Felka, explores the use of context-based prediction of network load to counteract congestion situations by means of transitions and ultimately improve service quality and customer satisfaction. Based on the preliminary work and in cooperation with Vodafone GmbH the transfer project aims to transfer the previous findings and approaches into practice and to deepen and further generalise them. The focus is on the development of analysis and forecasting models that are able to predict future network loads and, in particular, load peaks based on context factors and user-generated data from the Internet. Based on the prediction of network load, user preferences and customer satisfaction are also predicted in order to counteract network failures through countermeasures and to understand their influence on customer satisfaction and willingness to switch.

Learn more [tu-darmstadt.de]