Cost-Aware Cellular Networks Powered by Smart Grids and Energy Harvesting

Abstract

Introduction To meet the dramatic growth in wireless data traffic driven by the popularity of new mobile devices and mobile applications, the fifth generation (5G) of cellular technology has recently attracted a lot of research interest from both academia and industry (see, e.g., [1]). As compared with its fourth generation (4G) counterpart, 5G is expected to achieve a roughly 1000 times data rate increase via dense base station (BS) deployments and advanced physical layer communication techniques [1]. However, the large number of BSs will lead to large energy consumption and high electricity bills for cellular operators, which amounts to a large portion of their operational expenditure. For example, China Mobile owned around 920000 BSs in 2011 and the total energy cost per year was almost 3 billion US dollars, given that the annual cost for each BS is about 3000 US dollars [2]. Therefore, in the 5G era, it is becoming necessary for these cellular operators to reduce their energy costs by employing new cost-saving solutions in the design of cellular BSs, which are our main focus in this chapter. In general, these cost-saving solutions can be categorized into two classes, which manage the energy supply and the communication demand of cellular BSs, respectively [2–5]. On the supply side, one commonly adopted solution is to use energy harvesting devices (e.g., solar panels and wind turbines) at cellular BSs, which can harvest cheap and clean renewable energy to reduce or even substitute for the energy purchased from the grid [5]. However, since renewable energy is often randomly distributed in both time and space and cellular BSs are very energy-hungry, it is very difficult to solely use different BSs’ individually harvested energy to power their operation. As a result, the power grid is still needed to provide reliable energy to BSs. Besides serving as a reliable energy supply, the power grid also provides new opportunities for saving the BSs’ costs with its ongoing paradigm shift from the traditional grid to the smart grid. Unlike the traditional grid, which uses a one-way energy flow to deliver power from central generators to electricity users, the smart grid deploys smart meters at end users to enable both two-way information and two-way energy flows between the grid and the end users [6, 7].