Immersive Video Services at the Edge: an Energy-Aware Approach

To respond to the users’ demand for immersive and personalized media services, the 5G and the Multi-access Edge Computing initiatives are proposing novel network architectures. In this context, we present the Video Transcoding Unit (VTU) system, which exploiting the Cloud Enable Radio Access Network proposed by the EU 5G-PPP Sesame project, brings immersive video functionalities to the edge of networks, thus greatly improving User Experience with mobile terminals. A use case is discussed, in which the VTU is deployed in a Stadium or in a large public venue during a Crowded Event, to offer Immersive Video Services. In the proposed architecture, the VTU video processing component can be implemented as a Software-only Virtual Network Function running on different Hardware platforms (X86 or ARM architectures), eventually accelerated by a Graphics Processing Unit. Specific tests are described and discussed and specific Key Performance Indicators are introduced, showing the benefits of the Hardware-accelerated implementation, both in terms of computing performance and of energy efficiency. We believe that the proposed VTU framework significantly advances the state of the art in the provision of video services to the mobile users. Keywords-NFV; MEC; 5G; HW acceleration; GPU; Video

[1]  Jordi Pérez-Romero,et al.  Technology pillars in the architecture of future 5G mobile networks: NFV, MEC and SDN , 2017, Comput. Stand. Interfaces.

[2]  Jordi Pérez-Romero,et al.  Introducing Mobile Edge Computing Capabilities through Distributed 5G Cloud Enabled Small Cells , 2016, Mob. Networks Appl..

[3]  Wenwu Zhu,et al.  Multimedia Content Delivery with Network Function Virtualization: The Energy Perspective , 2017, IEEE MultiMedia.

[4]  Federico Pedersini,et al.  GPU-based VP8 encoding: Performance in native and virtualized environments , 2016, 2016 International Conference on Telecommunications and Multimedia (TEMU).

[5]  Mou Ling Dennis Wong,et al.  Distributed video transcoding on a heterogeneous computing platform , 2016, 2016 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS).

[6]  Federico Pedersini,et al.  Hardware-accelerated high-resolution video coding in Virtual Network Functions , 2016, 2016 European Conference on Networks and Communications (EuCNC).

[7]  Pietro Paglierani.  High Performance Computing and Network Function Virtualization: A major challenge towards network programmability , 2015, 2015 IEEE International Black Sea Conference on Communications and Networking (BlackSeaCom).

[8]  Claudio Meani,et al.  GPU-accelerated Video Transcoding Unit for Multi-access Edge Computing Scenarios , 2017, ICON 2017.

[9]  Guochu Shou,et al.  QoE analysis of NFV-based mobile edge computing video application , 2016, 2016 IEEE International Conference on Network Infrastructure and Digital Content (IC-NIDC).

[10]  Henning Schulzrinne,et al.  Real-Time Streaming Protocol Version 2.0 , 2016, RFC.