MU-MIMO User Grouping for WLANs supporting Indoor VR Applications

Abstract

Virtual Reality (VR) applications require low latency transmission of high-definition video frames to support immersive user experience without incurring side-effects like motion sickness. In this paper, we claim that state-of-the-art MU-MIMO user selection schemes fail to achieve the aforementioned goal due to their focus on building the best MU-MIMO group with maximal throughput. To fill the gap, we propose two new scheduling schemes to effectively reduce the user-experienced latency in WLAN-based VR systems. Via extensive simulations, we compare the performance of our proposed schedulers with that of Guidepost, a well-known MU-MIMO scheduler, to show the need of considering inter-group dependency in user scheduling to best support VR service provisioning. The latency enhancement by the proposed algorithms is revealed as up to 23.2% in the average sum airtime and up to 36.9% in the maximum sum airtime.