White Papers

ManualsBrandsDell ManualsConverged InfrastructureHigh Performance Computing Solution Resources

The Future of Mobile 5G Radio Networks and

Why Dell EMC is Looking Ahead

The Transition from 4G to 5G

With mobile technology often substituting as the modern day primary

computing resource, the demand for increased mobile services has

propelled mobile providers to recognize that 4G LTE mobile infrastructure

is no longer adequate. A traditional 4G LTE radio access network diagram

is shown below in Figure 1. At the heart of this is the baseband unit (BBU),

which provides the backhaul interface to the mobile network core and the

front haul interface to the remote radio head (RRH).

One of the obstacles to widespread deployment of new wireless networks

is the potential cost of customized equipment. Instead of utilizing standard

IT equipment, such as servers or switches, functions in these networks

have traditionally been performed by purpose-built devices. Using these

proprietary components eliminates a simplified path to increasing

performance at a fluid and scalable trajectory.

These insights served as a catalyst to the NFV (Network Functions

Virtualization) movement. The goal of NFV is to standardize the

telecommunications network infrastructure by steadily introducing an

ecosystem of server technology. One of several visions of NFV is to

implement the BBU functions using servers. As seen in Figure 2, a C-RAN

(Centralized Radio Access Network) can use a Centralized Unit (CU) and

distributed Units (DU) for baseband processing.

Tech Note by

Matt Ogle

Andy Butcher

Summary

Communication service

providers are envisioning

increased demand for

mobile services including

media and content delivery,

mobile gaming, virtual

reality and connected

vehicles. To satisfy this

emerging demand, the

buildout of a 5G cellular

infrastructure has

commenced.

This tech note explores how

computing platforms could

have an integral impact on

the future framework of the

mobile 5G cellular

infrastructure, as well as

how the confluence of

FPGA accelerator

technologies within Edge

servers would enhance

computing performance to

support these radio network

workloads.

Direct from Development

Figure 1: Traditional 4G LTE radio access network infrastructure

Direct from Development

Server and Infrastructure

Engineering

Summary of content (4 pages)

PAGE 1
Direct from Development Direct from Development Server and Infrastructure Engineering The Future of Mobile 5G Radio Networks and Why Dell EMC is Looking Ahead Tech Note by Matt Ogle Andy Butcher Summary Communication service providers are envisioning increased demand for mobile services including media and content delivery, mobile gaming, virtual reality and connected vehicles. To satisfy this emerging demand, the buildout of a 5G cellular infrastructure has commenced.
PAGE 2
Figure 2: Design concept for 5G; substituting the BBU with C-RAN, composed of a CU (centralized unit) server and multiple DUs (distributed units) Technical Value Propositions of NFV Transitioning from a traditional infrastructure to C-RAN would lower the total cost of ownership, improve system capacity utilization, and offer a path to performance improvement at the cell edge. Detailed value propositions explaining the primary design and use case variances between Figures 1 and 2 are listed below: 1.
PAGE 3
5. C-RAN enables Coordinated Multipoint Transmission (CoMP); a technique to increase cell edge coverage and throughput through scheduling. By providing connections to several base stations at once, data can be passed through the least-loaded base stations for better resource utilization. On the receiver side, performance can be increased by using several sites to selectively use the best signal, which can change rapidly due to multipath fading.
PAGE 4
As seen in Figure 4, system performance was predictably improved with the FPGA Turbo offload. By enabling system accelerators to perform some of the workload, the CPU experienced nearly half of the cycles required for uplink and downlink layer 1 PHY processing; this implies that the CPU is being utilized twice as much as before.