Intelligent RAN Slicing for Broadband Access in the 5G and Big Data Era

Abstract

As the 5G pace is quickening, enabled by rapid coverage rollout and availability of devices, the fixed broadband market is poised to become more competitive. We envision that 5G NS can emerge as a game changer that accelerates the development of FMC while counterintuitively opening up opportunities for fixed operators and the like in need of higher access network capability by leveraging NS-assisted FMC. In this article, we present an architectural framework to foster cross-industry infrastructure collaboration from a RAN slicing perspective with the aim of reducing network costs and improving rollout efficiency in the broadband ecosystem. Under this framework, RAN slicing can turn wireless access into an anchor service offering that can be exploited by not only existing mobile operators with licensed spectrum, but by a host of slice tenants including fixed wireless access startups, digital subscriber line operators, and converged operators. In the latter instances, independent RAN slices can be instantiated and granularly scaled to support these tenants who may not own mobile infrastructure or spectrum, while maximizing reuse of their live networks. Following a discussion on the market drivers and rationales for RAN slicing, we outline some RAN slicing options for different operators to realize resource isolation and allocation. Furthermore, a big data-driven, artificial-intelligence- powered RAN slicing and resource management framework is introduced for enabling RAN automation. Finally, several future challenges are identified.