Efficient resource allocation and scheduling mechanism of XR traffic

EXtended reality (XR) has become one of the most important 5G and 5G-Advance media applications in the industry. Specifically, XR is a broad term covering augmented reality, mixed reality and virtual reality. Although XR traffic has some similar characteristics with URLLC traffic, there are also some differences, such as multiple periodic data streams, variable packets sizes and so on. Therefore, based on the analysis of XR-specific traffic, the paper aims to investigate efficient resource allocation and scheduling mechanism of XR traffic. For instance, with the configured grant (CG) transmission, the latency of uplink (UL) XR data flows can be reduced. In order to improve the reliability, K-repeated CG configuration can be considered. To help the gNB make efficient scheduling, the UE can also use the assistance information to indicate XR traffics’ quality requirements and time-aligned transmission to gNB. After that, the paper analyses and evaluates the efficiency of the UL grants allocation method. It shows that with the assistance information, the gNB can make priority classification and efficient resource allocation, which can reduce resources by up to 81%.