Multiplexing Gain and Processing Savings of 5G Radio-Access-Network Functional Splits

Centralized radio access network (C-RAN) is a promising mobile network architecture designed to support the requirements of future 5G mobile networks. In C-RAN, the centralization of baseband units enables substantial savings of computational resources (what we call multiplexing gain in this paper) and significant power savings. On the other hand, the deployment of C-RAN requires high capacity and imposes strict latency requirements on the fronthaul transport-network. To address these issues, various alternative architectures, known as RAN functional splits, have been introduced to relax these strict fronthaul requirements. In this paper, we perform a quantitative analysis of the computational savings and the resulting power savings enabled by C-RAN, considering different RAN functional splits. To this end, we analytically model RAN computational resources to evaluate the multiplexing gain for different RAN functional splits. This model allows to calculate the processing reduction occurring in each RAN functional split. We then use this model to estimate the power savings of the various functional splits, considering different assumptions in terms of geographical areas, users distribution and number of aggregated cell sites. We find that up to 28% computational resources savings and 24% power savings can be achieved through functional splits in comparison to distributed RAN.