A Network Slice Resource Allocation Process in 5G Mobile Networks

Abstract

The fifth generation of mobile networks (5G) is associated with a wide spectrum of novel use cases that introduce a large number of very diverse requirements, regarding for instance throughput, latency, delay, availability and reliability. End-to-end network slicing is seen as a solution that allows to simultaneously accomplish those manifold requirements in isolated slices running on a shared network infrastructure. However, embedding those virtual end-to-end network slices into a common physical network containing wireless as well as wired network elements, while meeting all the different requirements, is still an unsolved problem. In this paper, a vision of an end-to-end network slice resource allocation process will be presented allowing to give fast feedback to a network operator or tenant on the feasibility of embedding new network slices. The associated research challenges will be discussed, especially focusing on the more complex Radio Access Network (RAN) resource allocation.

A Formal Optimization Model for 5G Mobile Network Slice Resource Allocation

“…Therefore, network management and planning need to be able to flexibly configure and modify NSLs as well as allocate required resources for the demanded network functions and services. As already explained in [3], this necessitates a high degree of automation of NSL preparation and deployment, as well as fast and efficient decision making on whether or not a Network Slice Embedding (NSE) is feasible. NSLs are intended to provide services to mobile and immobile end users.…”