A Statistical Characterization of SINR Coverage and Network Throughput with Macro-Diversity

In this paper, we consider a cellular network in which the user equipments (UEs) connect to multiple base stations (BSs) for augmenting their signal to interference plus noise ratio (SINR) coverage. In literature, the improvement in the SINR performance with multi-connectivity, or macrodiversity has been well-investigated for several network scenarios, e.g., millimeter wave (mm-wave) networks, networks with random blockages, etc. However, the impact of such multi-connectivity schemes on the network throughput has relatively sparse literature. We bridge this gap by jointly characterizing the effect of multi-connectivity on the SINR coverage and the throughput coverage of the UEs in the network, leveraging tools of stochastic geometry. Such a characterization for generic n connections to the typical user is necessary for 5G and beyond cellular networks, precisely since the degree of multi-connectivity should necessarily be dynamic and self-organizing in nature. Our results show that although the SINR coverage of the UEs always improves with more number of connections, the per-UE throughput in the network may increase or decrease depending on the density and data-rate requirement of the UEs. Interestingly, if the data-rate requirement of the UEs increases, our results show that the UEs should connect to a fewer number of BSs to maximize the per-UE throughput.