IEEE Access (Jan 2018)

LayBack: SDN Management of Multi-Access Edge Computing (MEC) for Network Access Services and Radio Resource Sharing

  • Prateek Shantharama,
  • Akhilesh S. Thyagaturu,
  • Nurullah Karakoc,
  • Lorenzo Ferrari,
  • Martin Reisslein,
  • Anna Scaglione

DOI
https://doi.org/10.1109/ACCESS.2018.2873984
Journal volume & issue
Vol. 6
pp. 57545 – 57561

Abstract

Read online

Existing radio access networks (RANs) allow only for very limited sharing of the communication and computation resources among wireless operators and heterogeneous wireless technologies. We introduce the LayBack architecture to facilitate communication and computation resource sharing among different wireless operators and technologies. LayBack organizes the RAN communication and multiaccess edge computing (MEC) resources into layers, including a devices layer, a radio node (enhanced Node B and access point) layer, and a gateway layer. LayBack positions the coordination point between the different operators and technologies just behind the gateways and thus consistently decouples the fronthaul from the backhaul. The coordination point is implemented through a software defined networking (SDN) switching layer that connects the gateways to the backhaul (core) network layer. A unifying SDN orchestrator implements an SDN-based management framework that centrally manages the fronthaul and backhaul communication and computation resources and coordinates the cooperation between different wireless operators and technologies. We illustrate the capabilities of the introduced LayBack architecture and SDN-based management framework through a case study on a novel fluid cloud RAN (CRAN) function split. The fluid CRAN function split partitions the RAN functions into function blocks that are flexibly assigned to MEC nodes, effectively implementing the RAN functions through network function virtualization. We find that for non-uniform call arrivals, the computation of the function blocks with resource sharing among operators increases a revenue rate measure by more than 25% compared to the conventional CRAN where each operator utilizes only its own resources.

Keywords