System Architecture of Fog Radio Access Networks

Abstract

As a promising architecture, the fog radio access network (F-RAN) can achieve high spectrum efficiency, energy efficiency, and low-latency by leveraging edge computing, cloud computing, and heterogenous networking. To fully unleash the potentials of F-RANs, in this chapter, a comprehensive overview of system architecture of F-RANs is provided. Specifically, the traditional system architecture of F-RANs is first introduced, which features a cloud computing layer, a network access layer, a logical fog layer, and a terminal layer. Then, network slicing in F-RANs for 5G is discussed to flexibly provide a soft-defined networking in a cost-efficient way. As for the radio interface, non-orthogonal multiple access, interference control, and transmission mode selection are identified as three key techniques. Later on, several application cases are demonstrated, including F-RAN enabled vehicles-to-anything communication, F-RAN enabled space communication, and so on. Finally, motivated by the recent advances in artificial intelligence (AI), F-RANs are anticipated to evolve towards an AI-driven paradigm with a glance at the corresponding architecture, principles, and open issues.