A Multi-Connectivity Architecture with Data Replication for XR Traffic in mmWave Networks

Abstract

mmWave communications are paving the way for next generation cellular networks due to their inherent ability to provide high data rates and mitigate interference. However, the intermittency of mmWave links and the advent of low-latency eXtended Reality(XR) applications presents a conundrum for satisfying strict QoS constraints. We propose a multi-tiered multi-connectivity network architecture which exploits mmWave macro-diversity to allow users (UEs) to connect to multiple base stations (gNBs) simultaneously and rapidly switch between them. The power of our proposed architecture lies in multiple tiers of multi-connectivity, wherein we selectively replicate UE data at a subset of associated gNBs in order to improve the response time to blockages. We evaluate the performance of XR traffic with standard scheduling algorithms. Our results show that connecting to multiple gNBs and enabling the scheduler to rapidly switch between them shields the UEs from higher handover delays and minimizes data plane interruptions. Although we show that our network architecture allows for much better performance even with conventional scheduling algorithms, we also highlight the need for better scheduling algorithms optimized for the multi-connectivity paradigm.