The Technique of Modelling and Statistical Analysis of Energy Consumption in 5G Multi-Tier Radio Access Networks

This paper notes that 5G networks, especially those with dense small cell deployment, suffer from high energy losses under low load conditions. The main reason is that the base stations are not able to properly configure the network configuration while the load is dynamically changing in time and space. For this purpose, we study in detail the Voronoi-based 5G network planning method, which is based on an algorithm for switching small-cell base stations on and off. The Voronyi tessellation allows optimizing delimitation of base stations. At the same time, the method is more energy efficient than other methods. The modeling studies carried out in this paper revealed high energy efficiency both for each instant and for network operation over time. For instance, over time about one quarter of energy can be saved. This value is relatively stable for different realizations. We also briefly overview our future work on improving 5G/6G radio-access network planning methods and energy consumption estimation techniques using artificial intelligence and Big Data technologies to improve energy saving.