Softwarization and Optimization for Sustainable Future Mobile Networks: A Survey

Due to the tremendous growth in mobile data traffic, cellular networks are witnessing architectural evolutions. Future cellular networks are expected to be extremely dense and complex systems, supporting a high variety of end devices (e.g., smartphone, sensors, machines) with very diverse QoS requirements. Such an amount of network and end-user devices will consume a high percentage of electricity from the power grid to operate, thus increasing the carbon footprint and the operational expenditures of mobile operators. Therefore, environmental and economical sustainability have been included in the roadmap toward a proper design of the next-generation cellular system. This paper focuses on softwarization paradigm, energy harvesting technologies, and optimization tools as enablers of future cellular networks for achieving diverse system requirements, including energy saving. This paper surveys the state-of-the-art literature embedding softwarization paradigm in densely deployed radio access network (RAN). In addition, the need for energy harvesting technologies in a densified RAN is provided with the review of the state-of-the-art proposals on the interaction between softwarization and energy harvesting technology. Moreover, the role of optimization tools, such as machine learning, in future RAN with densification paradigm is stated. We have classified the available literature that balances these three pillars, namely, softwarization, energy harvesting, and optimization with densification, being a common RAN deployment trend. Open issues that require further research efforts are also included.