Abstract: We are on a journey towards data- and AI-driven network and QoE management. Why is this happening now? Hardware, amount of available data, advances in AI algorithms and with 5G very low network latency and high throughput together enables this transformation. The transformation is also driven by pure need reduce the complexity of managing more and more complex networks and a multitude of applications and services using the network. AI is becoming essential for management of networks and to enable high-performance networks. Real-time AI, distributed AI, and combination of machine learning and reasoning are key components.
Bio: Jörgen Gustafsson is part of the Ericsson AI Research leadership, heading research teams in the areas of machine learning and artificial intelligence. He has experience from several Swedish national research projects together with academia and other industry companies. Jörgen is co-rapporteur of ITU-T Study Group 12 Question 14 since many years, and the technical focus is on AI, machine learning and Quality of Experience. Jörgen is an inventor of several patents and has a M.Sc. in computer science from Linköping University.
Abstract: the 5G Media Vertical Testbed is one of the federated facilities offered by the H2020 5GINFIRE project, specifically built to realize 5G support to media use cases. In this talk, I will introduce several functionalities of this facility, that enable the execution of media use cases beyond the state of the art of what is possible today. Specifically, we will look at two target areas: low latency VR streaming, and professional video production (uplink streaming) for live TV. These use cases are executed on two separate “Media Slices”, which are equipped with: (i) hardware and software functionalities to provide network assistance and media-specific in-network processing to media applications (e.g. by building upon and extending the recently published SAND standard from MPEG-DASH [ISO/IEC 23009-5:2017] and 3GP-DASH [3GPP TS 26.247, V15.1.0]), and (ii) 3rd party access APIs, to enable 3rd party service providers to monitor and manage the status and performance of their deployed applications on the 5G Media Vertical Testbed. The testbed employs a cloud infrastructure to provide easy deployment and scalability to services and (media slices).
Bio: Lucia received her PhD in 2012 from Delft University of Technology, the Netherlands, with a thesis on video streaming over peer-to-peer networks. She now works as a Senior Research Scientist at TNO, focusing on video distribution and on the impact of future internet architectures (e.g. ICN, SDN and 5G) on it. She usually takes the roles of project or technical lead for EU research projects and contract research projects that TNO conducts for several companies. Lucia also frequently advises European operators on network and TV technologies, contributes to international industry standards (especially in the 3GPP SA4 group, on behalf of Dutch operator KPN) and to 5GPPP and NEM visions on the 5G Media Vertical and pilots. Her research is regularly published in peer-reviewed academic venues and has led to more than 15 patents. Next to her researcher’s activities, Lucia is also responsible for managing TNO’s project portfolio for Immersive Human Communication (with a yearly revenue of about 2M euro).