A programmable and adaptive framework for 5G Network Slicing

The dawn of 5G brings with it the promise of significantly higher throughput and much lower latency delivered with high reliability by a softwarized and highly programmable network. 5G is quite different from the earlier wireless generations due to the huge diversity in the use cases it is aimed to cater to, including not-yet-foreseen services and requirements from new deployment scenarios. All of this makes a strong case for Network Slicing which logically, and in some cases, also physically partitions the network into slices, with each slice catering to a set of use cases (e.g., a slice each for enhanced mobile broadband (eMBB), massive machine-type communications (mMTC), and ultra-reliable low-latency communications (URLLC)), tenants, subscriber classes, industry verticals or a combination of one or more of these. Present network slicing techniques do not present an approach that is modular and flexible, so they have limitations in fitting into many deployment scenarios and/or have interoperability constraints. We propose a highly modular and flexible architecture in which components are of plug-and-play nature. We also propose an iterative and multi-step resource allocation strategy to optimally allocate resources to meet the slice and consequently the service Service Level Agreements (SLAs)/Key Performance Indicators (KPIs). Our proposed architecture and resource allocation approach can be applied to almost any kind of practical deployment scenario.