oa Green Techniques for Environment-Aware 5G Networks

Over the last decade, mobile communications have been witnessing an unprecedented rise of mobile user demand that is perpetually increasing due to the introduction of new services requiring extremely fast and reliable connectivity. Moreover, there is an important increase of the number of devices connected to cellular networks because of the emergence of the machine-type communication and internet of things. Indeed, data traffic on mobile networks is increasing at a rate of approximately 1.5 to 2 times a year, therefore mobile networks are expected to handle up to 1000 times more data traffic in 10 years time. Because of this huge number of wireless terminals, in addition to the deployed radio access networks (RANs) necessary to serve them, future fifth-generation (5G) cellular networks will suffer from an enormous growth of energy consumption that will cause negative economical and environmental impacts. It is predicted that if no actions are taken, the greenhouse gas (GHG) emissions per capita for ICT are estimated to increase from 100 kg in 2007 to about 130 kg in 2020. Therefore, there is an urgent obligation to develop new techniques and technologies in order to cope up with the exponential energy growth and correspondingly the carbon emission of emerging wireless networks. From a cellular network operator perspective, reducing fossil fuel consumption is not only for behaving green and responsible towards the environment, but also for solving an important economical issue that cellular operators are facing. Indeed, such energy consumption forces mobile operators to pay huge energy bills which actually constitute around the half of their operating expenditures (OPEX). It was shown that, currently, cellular networks consume around 120 TWh of electricity per year and mobile operators pay around 13 billion dollars to serve 5 billion connections per year.

Therefore, there is a growing necessity to develop more energy-efficient techniques to enhance their green performance while respecting the user's quality of experience. Although most of the proposed studies were focusing on individual physical layer power optimizations, more sophisticated and cost-effective technologies should be adopted to meet the green objective of 5G cellular networks. This study investigates three important techniques that could be exploited separately or together in order to enable the wireless operators achieve significant economic benefits and environmental savings:

- Cellular networks powered by the smart grid: Smart grid is widely seen as one of the most important means that enhance energy savings and help optimize some of consumers' green goals. It can considerably help in reducing GHG emissions by optimally controlling and adjusting the consumed energy. Moreover, it allows the massive integration of intermittent renewable sources and offers the possibility to deliver electricity in a more cost-effective way with active involvement of customers in the procurement decision. Therefore, introducing the concept of smart grid as a new tool for managing the energy procurement of cellular networks is considered as an important technological innovation that would significantly contribute to the reduction of mobile CO2 emissions.

- Base station sleeping strategy: Several studies show that over 70% of the power is consumed by base stations (BSs) or long term evolution eNodeB (LTE-eNB) for 4G networks. Turning off redundant or lightly loaded BSs during off-peak hours can contribute to the reduction of mobile network energy consumption and GHG emissions.

- Green networking collaboration among competitive mobile operators: The fundamental idea was to completely turn off the equipment of one service provider and serve the corresponding subscribers by infrastructure belonging to another operator. However, random collaboration may lead to the increase of certain mobile operator's profit at the expense of other competitive operators. This can cause a high energy consumption and a very low profit for the active network. Therefore, fairness criteria should be introduced for this type of collaboration.

In this study, we present in detail the techniques described above and provide multiple simulation results measuring the gain that could be obtained using these techniques compared to that of traditional scenarios.