Energy Management Strategies for Sustainable 5G Mobile Networks

The massive use of Information and Communications Technology (ICT) is increasing the amount of energy drained by the telecommunication infrastructure and its footprint on the environment. With the advent of the smartphone, mobile traffic is massively growing driven by both the rising number of user subscriptions and an increasing average data volume per subscription. This is putting a lot of pressure on the mobile network operators side, which are enforced to boost their infrastructure capacity by densifying the network with more Base Stations (BSs) and resources, which translates to a growth in the energy consumption and related costs. Hence, any future development in the ICT sector and its infrastructure has definitely to cope with their environmental and economical sustainability, where energy management is essential. In this thesis, we discuss the role of energy in the design of eco-friendly cost-effective sustainable mobile networks and, in particular, we elaborate on the use of Energy Harvesting (EH) hardware as a means to decrease the environmental footprint of the 5G network. Specifically, we investigate energy management strategies in 5G mobile networks with the main goals of: (i) improving the energy balance across base stations and other network elements, (ii) understanding how the energy can be exchanged either among network elements and the electrical grid, and (iii) investigating how renewable energy sources can be utilized within network elements to maximize the utility for the overall network in terms of better performance for the users (e.g., throughput, coverage, etc.), and lower energy consumption (i.e., carbon footprint) for the 5G network infrastructure. Therefore, we address, formulate and solve some of the problems related to the energy management in different scenarios within the 5G mobile network. The main covered topics are: (i) Wireless Energy Transfer where we investigate the tradeoffs involved in the recharging process from base stations to end users; (ii) Energy Cooperation in Mobile Networks where we target deployments featuring BSs with EH capabilities, i.e., equipped with solar panels and energy storage units, that are able to transfer energy among them; (iii) Energy Trading with the Electrical Grid where energy management schemes to diminish the cost incurred in the energy purchases from the electrical grid are pursued; and (iv) Energy Harvesting and Edge Computing Resource Management where EH and Mobile Edge Computing (MEC) paradigms are combined within a multi-operator infrastructure sharing scenario with the goal of maximizing the exploitation of the network resources while decreasing monetary costs. Online learning techniques, such as Gaussian Processes and Machine Learning Neural Networks, and adaptive control tools, like Model Predictive Control, are put together to tackle these challenges with remarkable results in decreasing costs related to energy purchases from the electrical grid and energy efficiency among network elements.