5G on the Horizon: Key Challenges for the Radio-Access Network

TLDR

5G will interconnect numerous devices and networks, driving ever‑increasing traffic and requiring heterogeneous deployments of diverse cell sizes and access technologies to meet stringent wireless network demands. The study explores how intelligence—via software‑defined networking and network function virtualization—can be integrated into heterogeneous cloud radio‑access networks to deliver energy‑ and cost‑efficient service quality. An overview is presented showing that applying intelligence to 5G RANs can satisfy quality‑of‑experience, energy, cost, and resource efficiency requirements.

Abstract

Toward the fifth generation (5G) of wireless/mobile broadband, numerous devices and networks will be interconnected and traffic demand will constantly rise. Heterogeneity will also be a feature that is expected to characterize the emerging wireless world, as mixed usage of cells of diverse sizes and access points with different characteristics and technologies in an operating environment are necessary. Wireless networks pose specific requirements that need to be fulfilled. In this respect, approaches for introducing intelligence will be investigated by the research community. Intelligence shall provide energy- and cost-efficient solutions at which a certain application/service/quality provision is achieved. Particularly, the introduction of intelligence in heterogeneous network deployments and the cloud radio-access network (RAN) is investigated. Finally, elaboration on emerging enabling technologies for applying intelligence will focus on the recent concepts of software-defined networking (SDN) and network function virtualization (NFV). This article provided an overview for delivering intelligence toward the 5G of wireless/mobile broadband by taking into account the complex context of operation and essential requirements such as QoE, energy efficiency, cost efficiency, and resource efficiency.