Near Real-Time Data-Driven Control of Virtual Reality Traffic in Open Radio Access Network

Abstract

In mobile networks, Open Radio Access Network (ORAN) provides a framework for implementing network slicing that interacts with the resources at the lower layers. Both monitoring and Radio Access Network (RAN) control is feasible for both 4G and 5G systems. In this work, we consider how data-driven resource allocation in a 4G context can enable adaptive slice allocation to steer the experienced latency of Virtual Reality (VR) traffic towards a requested latency. We develop an xApp for the near real-time RAN Intelligent Controller (RIC) that embeds a heuristic algorithm for latency control, aiming to: (1) maintain latency of a VR stream around a requested value; and (2) improve the available RAN allocation to offer higher bit rate to another user. We have experimentally demonstrated the proposed approach in an ORAN testbed. Our results show that the data-driven approach can dynamically follow the variation of the traffic load while satisfying the required latency. This results in 15.8% more resources to secondary users than a latency-equivalent static allocation.

IEEE GLOBECOM 2023

References in corpus (4)

• A Survey on Open Radio Access Networks: Challenges, Research Directions, and Open Source Approaches
• Estimation of Video Streaming KQIs for Radio Access Negotiation in Network Slicing Scenarios
• Measuring and Estimating Key Quality Indicators in Cloud Gaming services
• Temporal Characterization of VR Traffic for Network Slicing Requirement Definition