NOMA-Based Grant-Free Massive Access for Latency-Critical Internet of Things: A Scalable and Reliable Framework

From a system-level point of view, a grant-free random access protocol for latency critical internet of thing (IoT) must be designed to trade resource efficiency and target performance in a scalable manner. To this end, its performance characteristics must be fully understood by identifying the underlying physical-layer structure and constraints. In this article, we present a comprehensive structure of contention transmission unit for grant-free random access that employs a multi-signature spreading to average out multi-user interference subject to non-or-thogonal multiple access (NOMA). This structure is shown to trade preamble collision and activity detection failure optimally to achieve a target performance with the given physical resources at a varying level of user activity. Furthermore, as spectral efficiency and delay requirements are mainly governed by reliability of random access, we will discuss multiple receiver antenna-based approaches that play a crucial role in improving the reliability and supporting massive connectivity. The scalable and reliable features in all these aspects will become a useful part of the design framework for low latency and massive connectivity of 6G IoT applications.