Optimizing Cell Sizes for Ultra-Reliable Low-Latency Communications in 5G Wireless Networks

The millimeter-wave (mmWave) band with large antenna arrays and dense base station deployments has become the prime candidate for 5G mobile systems and key enabler for ultra-reliable low-latency communications (URLLC). In this paper, we propose an approach to estimating the optimal cell sizes of 5G networks that support URLLC services by combining both physical and data link layers, leveraging concepts from stochastic geometry and queuing theory. Furthermore, the impacts of the densification of base stations on the average blocking probability, which are of practical interest, are investigated with numerical results. The results show that the signal-to-noise-and-interference ratio (SINR) coverage probability and the average blocking probability achieve optimal values at different cell sizes. Moreover, the differences between the two types of optimal values become more significant with higher SINR thresholds. Our results suggest that traditional SINR-based approach for cell sizing will cause over-provisioning of base stations and significantly higher costs. Specifically, we share the insight that the interactions between SINR at physical layer and retransmission at link layer contribute to varying cost saving.