Assessing the Impact of Computational Resources to the Quality of Experience Provided by vRANs

Abstract

The new generation of mobile networks goes beyond classic radio-focus access networks by adopting the softwareisation of functions that typically relied on special-purpose hardware, now to be deployed in as general as possible purpose platforms. The resulting virtualised Radio Access Network (vRAN) follows the Network Function Virtualisation (NFV) paradigm, enabling a distributed and scalable network architecture. However, this approach increases the complexity of management tasks as computing resources now play an essential role in the network provisioning process. In this sense, this work aims to assess the impact of computational resources on the Quality of Experience (QoE) of video streaming services delivered over vRAN-based networks. This is done by performing multiple playback of 4K video while allocating varying numbers of Physical Resource Blocks (PRBs) and virtual CPU to vRAN. The obtained results prove that inadequate resource assignment to vRAN instances leads to degradation of the QoE, even if the allocation of radio resources is adequate for the service.

Keywords

• 5G
• computational resources
• mobile networks
• QoE
• video streaming
• vRAN