Experimental evaluation of virtual reality applications running on next-gen network scenarios with edge cloud assistance

Abstract

Volumetric video is an emerging key technology for immersive representation of 3D spaces and objects. Rendering volumetric video at client's end requires significant computational power which is challenging especially for mobile devices. One of the ways to mitigate this is to offload the rendering at the edge cloud and stream the video and audio to the thin client. Remote edge-cloud rendering may increase the end-to-end delay of the system due to the added network and processing latency which is greater than local rendering system. We investigate network latency in edge-based remote rendering over NextG networks and identify the bottleneck deteriorating the application performance. Further, we delve into the current state of the art and challenges of performing rendering remotely at the edge cloud and study the associated problems that need to be addressed in order to realize remote augmented reality(AR)/virtual reality (VR). Our prototype implementation shows effectiveness of maintaining the application QoE by prioritizing data at the level of a sub-flow and reducing the motion-to-photon latency.