The Implications of RAN Architecture Evolution for Transport Networks

Andy Sutton Principal Network Architect Network Strategy 25th June 2014

The Implications of RAN Architecture Evolution for Transport Networks

Contents

• Review current RAN architecture evolution trends

• Specify backhaul requirements to support LTE-A

• Consider the transport requirements for evolved D-RAN and new C-RAN solutions

• Present EE mobile backhaul strategy and target architecture for a future proof Het-Net

2

RAN and backhaul evolution

1. GSM to GSM/UMTS to GSM/UMTS/LTE

2. Multiple base stations to multi-RAT platforms

3. TDM, ATM and IP Transport Network Layers to all-IP TNL

4. Underlying transmission changes from TDM (PDH/SDH) to Carrier Ethernet

5. Transmission migration supported by PWE3 solutions

6. Denser networks with different site types – macro – micro – pico

7. VoIP and QoS differentiation 3

Current multi-RAT mobile backhaul network architecture

4

RAN architecture evolution trends 1. Increased peak and average speeds

2. Increased capacity

3. Reduced latency

4. Differentiated quality of service

5. Multiple radio access technologies

6. Multi-band operation

7. Increased coordination – scheduling

8. Distributed and/or Cloud RAN

9. Denser networks

10. Heterogeneous networks

11. Self organising/optimising networks

12. Time/phase alignment between cells

5

LTE-Advanced

• Carrier Aggregation - n x Component Carriers (CC) - 20+20MHz deployed

• Capacity and performance

• Heterogeneous Networks (Het-Nets)

• Public access small cells - radio functionality, fixed and wireless backhaul

• Coordinated Multi-Point Transmission and Reception (CoMP)

• Different techniques place different demands on backhaul

• Some techniques probably require C-RAN for effective implementation (fronthaul)

6

LTE-Advanced backhaul

• Capacity upgrades

• Dark fibre, Ethernet Leased Lines, Microwave radio etc.

• Enhancing performance - flatten the protocol stack…

• Minimising latency and frame/packet delay variation

• Minimising frame/packet error loss rates

• Flexible QoS to support multi-service

• VoLTE timeline aligns with new LTE-A features

• New connectivity to support advanced RAN features

• Coordinated scheduling - centralised management function

• Small cell backhaul - connectivity, performance and integration

• Flexibility to optimise traffic flows to support CoMP techniques

• Time and timing

• Frequency and phase (time of day) sync

7

D-RAN and C-RAN…

8

• Distributed RAN is an evolution of today’s base station architecture with improved coordination and scheduling capabilities

• Centralised scheduling

• Cloud RAN is a new functional split with centralised baseband (digital unit) coordinating across multiple remote radio units (radio unit)

• Introduction of fronthaul segment

• Benefits of C-RAN over D-RAN appear to be limited at the moment and therefore don’t justify significant investment in fronthaul

• Consider TCO based on circumstances

Target architecture and evolution strategy

9

• Focus on capacity and performance

• Short-medium term scalability of Ethernet leased lines and microwave radio systems

• Development of Super Macro sites in support of 2G, 3G and 4G national coverage and capacity

• Includes CA on macro’s, moving from 2 to 3 CC

• Introduction of optimised Het-Nets to enhance capacity and quality of experience

• Longer term partnerships with fixed providers who can offer fibre-like performance rather than complex Ethernet leased line products

• Introduction of innovative wireless backhaul technologies, new techniques, new bands etc.

Summary

10

• Mobile backhaul evolution is essential to support RAN and product evolution

• Higher throughput with improved performance is an enabler for new services

• We must decouple any relationship between capacity, cost and performance

http://www.thinksmallcell.com/Femtocell-Interview/thinksmallcell-interview-with-andy-sutton-principal-network-architect-at-ee-uk-on-their-future-small-cell-plans.html

THANK YOU

andy.sutton@ee.co.uk