Small Cell Advances Within Network Virtualization

Small Cell Advances Within Network Virtualization Small Cell Forum February, 2015 Clare Somerville SCF Virtualization Use Case Lead

Agenda 1. Small Cell Forum Introduction 2. Virtualized Small Cell Drivers 3. Virtualized/Cloudified Small Cells 4. Summary

1. Small Cell Forum Introduction and Work Program

The Small Cell Forum To accelerate small cell adoption to change the shape of mobile networks and maximize the potential of the mobile internet Not-for-profit, founded in 2007 Independent, inclusive, international Aims Ecosystem development Market education Driving open standards 62 operators covering 3 billion global mobile subscribers 44% of total 78 providers of small cell technology representing all parts of the ecosystem

Small Cell Forum Release Planning

Release Roadmap Each release addresses the full range of small cell use cases, and will provide a major update to one or more packages along the lines of a theme

3. Virtualized Small Cell Drivers

2 BIG Innovations revolutionizing the design and economics of mobile networks

Why Virtualization matters to the Small Cell Forum 1. Cloud-RAN and small cells have often been presented as alternatives. We believe that virtualization techniques can be re-applied to small cell architectures. 2. Virtualization can be used to enhance co-operation within clusters of small cells. 3. Small cell clusters may be required to interwork with a virtualized macro layer. 4. In dense HetNets, small cells will deliver maximum ROI if they are fully interworking with other platforms to create the broadest pool of capacity and most efficient use of shared resources. 5. Our aim is to provide leadership in creating a unified view of the evolving role of virtualized architectures for dense HetNet.

Who s Planning What

Key Questions from the SCF s Operator Group 1. Is there any divergence between how virtualization is viewed from a macro RAN perspective versus a small cell perspective? 2. What are the real business benefits of virtualization in the RAN? Do benefits apply equally to macro and small cell deployment use cases? 3. Are current small cell architectures well suited to virtualization/cloudification? 4. Can virtualization deliver a phased roadmap, e.g., aligned with longer-term 5G directions 5. How do we ensure virtualization supports innovation across a multi-vendor ecosystem?

4. Virtualized/Cloudified Small Cells

Network Function Virtualization Use Cases #1: Network Functions Virtualisation-as-a-Service #2: Virtual Network Functions-as-a-Service #3: Virtual Network Platform-as-a-Service #4: VNF Forwarding Graphs #5: Virtualisation of Mobile Core Network and IMS #6: Virtualisation of Mobile Base Station #7: Virtualisation of the Home Environment #8: Virtualisations of CDNs #9: Fixed Access Network Functions Virtualisation Use Case #5: Release 5 focus on virtualizing small cell core network components Use Case #4: Release 5 focus on Stitching together components, e.g., security gateways and small cell gateways Use Case #6: Release 6 study into small cell virtualization

Initial stage to virtualization is to decompose into physical and virtual Remote Small Cell Centralized Small Cell Virtualized Function Remote Small Cell Remote Small Cell Virtual Network Function (VNF) Physical Network Function (PNF)

Then overlay PNF and VNF onto the ETSI NFV Architecture Framework

Various different decompositions possible SON OAM APPS RRC PDCP RLC Upper Lower Upper Lower RF Baseline Macro: CPRI based decomposition RRH PNF

Current observations regarding parallel Cloud RAN investigations No new decompositions are being standardized i.e., Cloud-RAN is based on CPRI or Compressed CPRI Bandwidth requires lambda or OTN service with very tight bounds on latency Incompatible with current GbE packet switched last mile networks that are preferred approach for deploying today s small cells Will decomposition based on (compressed) CPRI deliver on aims of Small Cell Forum? Accelerate small cell adoption - seems requirement for Lambda/OTN service to each small cell site will likely hinder mass adoption Change the shape of mobile networks seems there is no revolution as the innovation in the small cell ecosystem will be have to continually address the centralization versus distribution issue (and which may consequently hinder small cell adoption)

Various different decompositions possible SON OAM APPS RRC PDCP RLC Upper Lower Upper Lower RF Baseline Macro: CPRI based decomposition RRH PNF Split based decomposition PNF / based decomposition (FAPI) PNF Split based decomposition PNF RLC/ based decomposition PNF PDCP/RLC based VNF ( Xn ) Services VNF Distributed enb PNF

Mapping of Functional Splits/Delay Characteristics to Performance PDCP RLC RF PDCP RLC RF PDCP RLC RF PDCP RLC RF PDCP RLC RF CPRI Split / Split PDCP/RLC IDEAL DELAY GOOD DELAY NON-IDEAL DELAY BEST PERF. IMPROVEMENT LEAST PERF. IMPROVEMENT

5. Summary

Summary Network Function Virtualization will be increasingly applied to Mobile Networks: ETIS NFV Use Case #4 and #5 Small Cell Forum already leveraging Virtualized 3GPP Core Network (Use Case #5) to lower the barriers to entry for small cell deployment Critical aspect of Virtualized Base Station (Use Case #6 ) is the decomposition of the base station and its consequential requirements on the transport network SCF is performing a comprehensive analysis looking at all possible splits in order to understand the linkages between Small Cell virtualization, business cases, transport characteristics, deployment related issues Look out for the Publication at Small Cells World Summit - June 15

www.smallcellforum.org