When the network operator becomes a utility company: accounting for energy cost in wireless network design

Abstract

Nowadays, it is clear that society cannot live without telecommunications anymore. Fields like Internet of Things (IoT), Machine Learning (ML) or Artificial Intelligence (AI) are becoming essential for many applications and the number of devices connected to the Internet now outnumbers the number of human beings on planet Earth. Wireless Access Networks are the backbone of telecommunications and it is not an easy task to satisfy all the growing traffic demand caused by the increasing amount of users and devices in recent years. In addition, it is necessary to provide the energy required by these base stations that are part of the Wireless Access Networks. This energy needs to come from renewable and decarbonised sources, in order to reduce CO2 emissions and achieve an energy transition from fossil fuels to environmentally-friendly energy sources. In this thesis, solar, wind and geothermal energy are considered to feed a real scenario composed by a set of macrocells and microcells located in the city of Ghent. The algorithm developed for this purpose is combining energy and cost saving strategies to improve both energy efficiency and total economic costs of the network. The study proves that feeding networks with a mixed combination of renewable energy sources can save energy and at the same time provide benefits if a buying and selling energy approach is considered.