Sensing-Based Grant-Free Scheduling for Ultra Reliable Low Latency and Deterministic Beyond 5G Networks

Abstract

5G and beyond networks should efficiently support services with stringent and diverse QoS requirements. This includes services for verticals that demand Ultra Reliable and Low Latency Communications (URLLC). Scheduling strongly impacts the communication latency, and 5G NR introduces grant-free scheduling to reduce the latency at the radio level. Grant-free scheduling can use shared resources and the transmission of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">K</i> replicas per packet to increase the packet delivery ratio and efficiently utilize the spectrum. Previous studies have shown that existing 5G NR grant-free scheduling has limitations to sustain URLLC requirements for aperiodic (or uncertain) and deterministic traffic that is characteristic of verticals such as Industry 4.0 or manufacturing. In this context, this paper proposes and evaluates a novel grant-free scheduling scheme that can efficiently support deterministic and aperiodic uplink traffic. The scheme avoids packet collisions among UEs sharing resources using a priority-based contention resolution process that relies on the transmission of announcement messages in minislots and a local channel sensing process. This study demonstrates that the proposed sensing-based grant-free scheduling scheme outperforms current 5G NR grant-free scheduling implementations, and can support a higher number of UEs with URLLC and deterministic requirements with a considerably lower number of radio resources.