NetXRate: O-RAN enabled network assisted rate control for XR services
Alejandro Lopez Garcia, Amr A. AbdelNabi, Daniel Camps‐Mur, Miguel Catalan-Cid, Mario Montagud · 2024
Next-generation networks have become key enablers for distributed eXtended Reality (XR) services, which pose stringent demands in terms of bandwidth, compute and reliability. In this context, the Network as a Service (NaaS) paradigm, where the network offers APIs tailored to specific vertical domains, is a promising approach to enhance delivery protocols and communications for XR services. In this paper, we present NetXRate which is a new NaaS API tailored to interactive XR services. NetXRate leverages Open RAN (O-RAN) and the network exposure capabilities of the 5GCore, to deliver rate recommendations to XR applications based on the instantaneous capacity available in the cellular network. Using a simulative approach, we demonstrate how NetXRate can be integrated with two state-of-the-art Over-the-Top (OTT) XR rate recommendation strategies, achieving a reduction of up to 72% in outage probability in scenarios with severe network congestion where multiple XR users compete for higher share of bandwidth in the same cell.