Resource Allocation for Multi-User Downlink MISO OFDMA-URLLC Systems

TL;DR

This paper develops and compares optimal and low-complexity resource allocation algorithms for MISO OFDMA-URLLC systems, emphasizing the importance of multiple antennas and short packet transmission to meet strict latency and reliability requirements.

Contribution

It introduces a monotonic optimization-based optimal resource allocation method and a practical low-complexity algorithm tailored for URLLC constraints in MISO OFDMA systems.

Findings

Multiple antennas significantly reduce latency and improve reliability.

The sub-optimal algorithm closely matches the optimal performance.

Traditional Shannon-based resource allocation may violate delay constraints.

Abstract

This paper considers the resource allocation algorithm design for downlink multiple-input single-output (MISO) orthogonal frequency division multiple access (OFDMA) ultra-reliable low latency communication (URLLC) systems. To meet the stringent delay requirements of URLLC, short packet transmission is adopted and taken into account for resource allocation algorithm design. The resource allocation is optimized for maximization of the weighted system sum throughput subject to quality-of-service (QoS) constraints regarding the URLLC users' number of transmitted bits, packet error probability, and delay. Despite the non-convexity of the resulting optimization problem, the optimal solution is found via monotonic optimization. The corresponding optimal resource allocation policy can serve as a performance upper bound for sub-optimal low-complexity solutions. We develop such a low-complexity resource allocation algorithm to strike a balance between performance and complexity. Our simulation results reveal the importance of using multiple antennas for reducing the latency and improving the reliability of URLLC systems. Moreover, the proposed sub-optimal algorithm is shown to closely approach the performance of the proposed optimal algorithm and outperforms two baseline schemes by a considerable margin, especially when the users have heterogeneous delay requirements. Finally, conventional resource allocation designs based on Shannon's capacity formula are shown to be not applicable in MISO OFDMA-URLLC systems as they may violate the users' delay constraints.