Delay Analysis of Multi-User URLLC Systems Under Finite Block-Length Regime

This paper presents a framework designed to evaluate delay violation probabilities in multi-user down-link Ultra-Reliable Low Latency Communication (URLLC) systems. Emphasizing the significance of employing a finite block-length (FBL) transmission model to address transmission errors, our proposed framework assumes infinite user queues at the base station (BS). Delay violation probabilities are assessed with respect to predefined target delays, computed based on infinite queues. To provide insights into essential tradeoffs and performance limits, we conduct a comparative analysis of delay violation performances for three widely recognized user scheduling algorithms: round robin (RR), max-cqi, and longest queue (LQ), across diverse network conditions. Simulation results underscore the critical importance of choosing an appropriate scheduling algorithm to effectively meet demanding target delay violation probabilities.