Small Cell Open RAN:

A catalyst for new 5G

business models
An SCF guide to small cell Open RAN
architecture and deployment
Executive Summary
Open RAN is an important step toward the democratization of radio
access networks (RANs). Essentially, this approach aims to lower
costs, increase flexibility and innovation by enabling the equipment
of multiple stakeholders to be part of radio access networks through
open and fully-defined standards for interoperability.

The Open RAN concept itself is not new and has for some time
been explicit in many of today’s cellular standards. However, over
the years and for a variety of reasons, equipment vendors have
implemented the standards with proprietary enhancements, resulting
in what is essentially a closed radio access ecosystem dominated by
a small number of major players. Open RAN, in its current drive,
aims to fix this situation by disaggregating the radio access network
into meaningful functional parts and developing fully specified, open
interconnecting standards.

The suitability of such equipment is likely to be validated through

interoperability plugfests and certification programs. This approach
is being developed by a number of trade bodies – including Small Cell
Forum, which has already made significant contributions to the Open
RAN toolbox to meet requirements associated with a diverse set of
deployment and operational scenarios.

In fact, SCF has pioneered this approach since its foundation, by

nurturing a vibrant ecosystem of established and emerging component
and equipment vendors and developing mission-critical interface
standards. For example, SCF pioneered the development of IuH
interface (between small cells and the core network), small cell
management solutions (TR69/196) and FAPI specifications (between
the physical-layer and protocol-layer software).

Today, SCF members are building on this tradition by evolving and

enhancing the current small cell Open RAN specifications and
creating new ones for the 5G world, as well as enabling the expanding
ecosystem of small cell deployers and operators.

This brief report aims to provide a crisp and accessible overview of

Open RAN – what it is and what it’s not, along with a short overview
of SCF activities in this space.

And, finally, the report provides important market research data,

shedding light on how various Open RAN solutions are addressing
diverse usage, deployment and operational scenarios.

Prabhakar Chitrapu PhD

Chair, Small Cell Forum

2
Contents
Context: Evolution of Open RAN 4

Why is Open RAN so important now? 7

Open RAN requires a diversity of platforms 12

SCF Open RAN initiatives 16

Open RAN case studies from SCF members 20

Conclusions 27

SCF239.10.01 April 2022

3
Context: Evolution of Open RAN
The development of Open RAN architectures is one of the most discussed topics in the mobile
networks industry, but also one of the most misunderstood. Sometimes obscured by a great
deal of hype, politics and flawed assumptions, this is nevertheless a subject associated with a
very significant objective – to ensure that all the hardware and software elements of a radio
access network (RAN) can interoperate, even when they come from different suppliers.

And if this objective was achieved it would drive several important changes in the 2020s. It would:

• enable established operators to access broader bases of innovation and avoid supplier lock-in;

• lower barriers to new deployers, especially in greenfield scenarios such as large enterprises,
by driving a broad ecosystem of equipment at all price and performance points;

• lower barriers for new suppliers of equipment, software, chips and services to enter the
market and be competitive with established vendors.

A fully open framework, then, could transform the economics of mobile radio network deployment,
but there are a couple of key myths about Open RAN. One is that it is something new.

Myth 1. Open RAN is new

Open RAN is not a new concept, in theory, although it is poised to be new in practice
(i.e. deployment). In general, mobile networks have been based on open standards since
the definition of GSM for the 2G era, and the 3GPP continues to define interoperable
specifications for the entire network. Aside from the interface between the RAN and devices,
the most successful specs have been the interfaces between the RAN and core network, which
enabled suppliers to focus on one or the other, and deployers to independently deploy and
scale the two parts of the network.

However, the radio access network itself, although having some open interfaces, has been a
tightly integrated solution so far, employing interfaces that have vendor specific optimizations
and enhancements. The practical implication of this has been that it was not possible to mix
and match different parts of the RAN. Open RAN seeks to address that problem, so different
parts of the RAN have fully specified open interfaces, and one can indeed mix and match
parts from different vendors. The Open RAN initiative has been further strengthened by re-
architecting the RAN, so that there is further disaggregation of various network functions as
well as hardware and software.

4
Specifically, the data and control plane functions of the RAN are split into radio unit (RU), distributed
unit (DU) and centralized unit (CU), whereas the management functions are disaggregated into real-
time and non-realtime RAN Intelligent Controller (RIC). Similarly, the separation of the software
and hardware functions enables virtualized network functions, implemented on COTS hardware. Of
the RU, DU and CU, the CU is expected to be always virtualized, whereas DU may be virtualized
depending on the specific network dimensions (though requiring optimized hardware in the upper
PHY), and RU is generally not virtualized but rather realized as a physical network function (PNF).
Among the many advantages of virtulaization, the VNFs can be implemented in the cloud, which may
be a private cloud or a public cloud, depending on latency, security and privacy requirements.

The graphic below illustrates the concept of Open RAN, contrasting it with traditional RAN
and virtualized RAN (which separates hardware and software by open interfaces, but does not
necessarily allow mixing and matching of the RU, DU and CU).

Traditional RAN Virtualized RAN Open RAN

(vRAN)

Antenna

RF RF RF

Radio Unit

CPRI (closed) CPRI or eCPRI (closed) Front-haul (open)

Baseband

Distributed Unit

Backhaul (closed) Mid-haul (closed) Mid-haul (open)

Edge server Centralised Unit, Centralised Unit,

collocated with collocated with
Proprietary hardware and Edge server Edge server
software, single vendor
Off-the-shelf hardware, Off-the-shelf hardware,
proprietary interfaces open interfaces and
and software, also proprietary software,
Core Network single vendor multi-vendor

The connections between the CU and DU (called midhaul) and the RU and DU (called
fronthaul) have been a key focus of Open RAN efforts because if these are not truly open and
fully specified, it will be impossible to select an RU from one vendor to run with a CU or DU
from another. Small Cell Forum has specified one of the key solutions for open fronthaul, its
nFAPI interface, branded as Open6.

5
Myth 2. Open RAN is a single solution
A second myth is that Open RAN is a single solution. One of the reasons for the current
push behind open frameworks is, in the 5G era, RAN solutions are looking increasingly
diverse. Nationwide macro networks are joined by dense urban hotzones, by networks
to support transportation systems or industrial complexes, or by indoor networks to
enable new applications in enterprise scenarios. It is clear all these different networks
will not be optimally served by a single architecture or a single base station form factor,
and so development of open platforms is being driven by several initiatives, though with
considerable cooperation between them.

Small Cell Forum (SCF) has pioneered in the definition and promotion of open interfaces in
the RAN, with its flagship product, namely the FAPI interface (SCF222). Standing for an open
interface that was developed by a representative sample of the small cell ecosystem, it defines
an open interface between the physical layer (PHY) hardware and upper layer (MAC) software.
This open interface has become the defacto standard for small cell products worldwide and is
beginning to be adopted within the macro RAN network as well.

Other bodies, notably O-RAN Alliance, are focusing more heavily on achieving that open
macro outdoor network architecture, while the Open Networking Foundation is developing
RAN control standards called SD-RAN, and Telecom Infra Project is focused on adapting and
trialling raw specifications to align with real world operator requirements.

Small cell open RAN is a topic of intense interest and many of the efforts of SCF and its
members over the past decade, to create a common interoperable platform, are now being
replicated throughout the industry.

6
Why is Open RAN so important now?
As we have seen, there have always been initiatives to make RANs as open and multivendor-
capable as possible. Some have led to permanent standards being established and widely
supported, including the 3GPP specifications and SCF’s Iuh interface, which was adopted by
3GPP. Others have been agreed to but then implemented in non-standard, non-compliant ways
by each vendor. An example is CPRI, which established a common fronthaul interface between
remote radio units and baseband units, but was not implemented uniformly, with vendors being
unwilling to open up their radio technology to competitors.

There is the risk that something similar will happen with this generation of Open RAN
standards, but there is significantly greater momentum behind a fully open framework
this time around.

• Operators face a much-reduced choice of vendors after years of consolidation among base
station and small cell suppliers.

• There is government and international support for Open RAN, not just related to geopolitics,
but also for the chance for some countries or regions, such as the USA or India, to build local
mobile networking industries with benefits for the local economy and global influence.

• In the small cell environment in particular, there is a rising diversity of deployers to address
the varying needs of different industries. Many of these have very different economics from
those of the traditional mobile network operator (MNO), and want access to a flexible
ecosystem that will produce a range of solutions for different scenarios.

Although the Open RAN debate is often focused on cost reduction, the reasons for operators
(old and new) to throw their collective weight behind the emerging frameworks are far more
complex. The wide range of drivers for deployment of small cell Open RANs, as seen in Figure 1,
demonstrates why there is more urgency to make open architectures happen this time around.

Figure 1 is a result from a survey of 62 MNOs and 36 other small cell deployers, conducted by
SCF and Rethink Technology Research in August 2021. The respondents were asked to name
all the drivers to invest in small cell Open RAN between 2021 and 2025, and of the most-cited
drivers, they were then asked to name the three most important for their business model. The
results were also compared with a similar survey, conducted a year earlier.

7
 Figure 1. Drivers to support small cell Open RAN before 2025 (percentage of
operators placing each driver in their top 3)

50
45 2020 2021
40
% placing in top 3

35
30
25
20
15
10
5
0
simplify deployment
Accelerate and

of innovation
Access wider base

and rural...
Facilitate enterprise

interoperability
Ensure multivendor

Reduce TCO

competition
Drive price

virtualisation

dead ends
Avoid technology

of technology...
Take back control
Although reducing total cost of ownership (TCO) is always a priority for any deployer, it was Facilitate end-to-end
less prominent than several other considerations in the survey results. Many operators believe
cost reductions will be slower to materialize than some other gains, or they value benefits such
as simplified deployment (which will have an effect on cost too) more highly.

Key drivers for Open RAN

• Deployments enabled by mix-and-match architecture and common reference designs
emerged as the most important reason to support Open RAN, followed by the ability to
access a wider base of innovation in all elements of the network, including its integration and
services. These factors could be significant in accelerating roll-outs, and enable deployers to
adjust RAN architectures more flexibly to the different form factors and cost bases required
in different environments, from enterprise to rural.

• Another driver is to make it easier to deploy virtualized networks for the first time, since
significant amounts of small cell vRAN development is now focused on open architectures,
accelerating the process. The adoption of open architectures will often go hand-in-hand with
virtualization of the network.

Many of the factors highlighted in the survey have only come to the front of operators’
minds in recent years, as they face a perfect storm of new cloud-based architectures, new
deployment scenarios (especially in enterprise) and the migration to 5G. If open architectures
can make these parallel and challenging changes easier and less risky - as more than two-thirds
of deployers believe they can, according to our study – it is clear why Open RAN is seen as
urgent, rather than nice to have, in the 2020s.

8
The rise of demand in enterprise environments
Rising demand for cellular networks in enterprise environments - to support high quality in-
building connectivity, or to enable new processes and use cases in industrial campuses – is
particularly important in making Open RAN a necessity now. It is also the reason why we
expect early Open RAN deployments to be led by small cells:

• most enterprise networks will be localized, by their nature, and so reliant on small cells; and

• many will be rolled out and operated by non-traditional providers such as private network
operators or enterprise integrators and neutral hosts. These will be better positioned than
established MNOs with large installed networks to quickly adopt new architectures and
potentially new vendors.

Figure 2 shows our forecast for the deployment of small cells supporting Open RAN
architectures, and by 2026 we expect open systems to account for 80% of total new small
cells deployed, a far higher figure than the 45-50% predicted in the macro network. This
progress is partly because of the lower barriers to change in these relatively greenfield cellular
environments and the rising intensity of demand for new networks, which may often start with
4G before moving to 5G. All this effort in small cell networks will help to create a body of
experience and expertise that will have valuable lessons for other types of RAN also, in time.

Figure 2. Small cells supporting open RAN architectures by environment, and

percentage of total deployments that support these architectures

3,500 90
80
3,000
70
2,500
60
,000 RU

2,000 50

1,500 40
30
1,000
20
500
10
0 0
Year 2021 2022 2023 2024 2025 2026
Enterprise 370 751 1,663 2,171 2,858 3,151
Urban and rural 43 148 460 806 1,112 1,518
% open RAN 12 23 44 56 70 77

9
Open RAN at the heart of
Cellnex Innovation
Open RAN is a new architecture that disaggregates the different
elements of the Radio Access Network (RAN) under open interfaces
thus enabling greater choice and flexibility for mobile players when
rolling out a network.

Thanks to this disaggregation Open RAN aims to drive innovation,

expansion of the supplier ecosystem and cost savings.

Open RAN offers Cellnex the opportunity to extend its portfolio

of services through “Adjacent Assets ”(fibre optics, mobile edge
computing and data centres), services that Cellnex is already providing,
as Open RAN would facilitate the sharing of the radio because of the
standardization of the active equipment.

Additionally, Cellnex will be able to complement the service currently

provided in the field of passive infrastructure with General Purpose
Hardware (GPH), both as Infrastructure as a Service or Platform as a
Service for the Open RAN.

Open RAN is therefore a strategic opportunity for Cellnex that will

allow to strengthen and tighten its relationship with customers by
expanding throughout the value chain.

Cellnex Telecom is currently running different Innovation projects

with the aim of evaluating the advantages of Open RAN in
particular for the deployment of private networks over 5G,
multitenant small cells and Edge Computing.

10
Increasing innovation and choice
We fully support the emergence of Open RAN as it opens up the
value chain, creating a new ecosystem and increasing innovation and
choice for ourselves and our customers.

With the mobile network operators (MNOs) understandably focusing

on more pressing macro-cellular commitments, architecturally speaking
they’ll be looking to experiment with Open RAN. Therefore, the more
immediate opportunities for the nascent Open RAN industry and
potential new vendors lie in small cell, in-building and campus solutions.

Freshwave has long championed the neutral host approach to in-

building and campus connectivity. We’ve connected more buildings
to multiple operators, including all four of the UK operators than
any other neutral host provider. And we believe that open RAN
will eventually bring a more varied vendor landscape with a smaller
footprint and, crucially, a cheaper product set that we can match
better to the physical building, campus and customer requirements.
I say eventually, as there is still some way to go before all these
potential benefits will be fully realised.

The economics and customer focus of any commercial property

provider makes neutral host the natural choice and gives tenants the
most possible flexibility. And the extension of Open RAN to provide
landlords and tenants with a mobile private network is a natural next
step for Freshwave’s services. However, moving into this space could
also bring with it the need to support critical applications, so any Open
RAN provider needs to ensure that they can guarantee their product is
highly available, including potential service level commitments.

Freshwave is already leading Open RAN work and we look forward

to working with the industry at large, including DCMS and the
SmartRAN Open Network Interoperability Centre, to develop this
technology further for the benefit of the industry and our customers.

Tom Bennett,
CTO, Freshwave

11
Open RAN requires a diversity of platforms
We have seen that one important driver for small cell Open RAN is to enable a broad and
varied ecosystem. Open standardized interfaces help specialist or new vendors to reduce their
cost and time to market because many specifications and APIs are standard, and it enables
operators to select smaller suppliers with reduced risk. All this is vital in enterprise markets
where no single solution will fit all scenarios.

In any connectivity market, there is a fine balance between uniformity, which encourages scale
and allows many companies to innovate on a common platform; and diversity, to support a wide
range of solutions for different scenarios. In Open RAN, our surveys of operators and enterprise
deployers show clearly that a single architecture is unlikely to support all the possible solutions
optimally, but too much fragmentation will slow progress and reduce cost efficiency.

The consensus is that two or three architecture options will be needed to avoid a ‘lowest
common denominator’, but this number would be sufficiently small to allow scale to be built in
each architecture. Moreover, operators are clear that all Open RAN candidates should work
within even broader frameworks, such as those developed by ETSI and 3GPP, and cooperate to
allow for interworking.

SCF has a long history of cooperating with other industry alliances and standards bodies to
align its own work, which is optimized for key small cell use cases, with initiatives that address
other aspects of mobile communications. It has partnerships and cooperations with 3GPP,
ETSI, Wireless Broadband Alliance, O-RAN Alliance, Open Air Interface and many others

One of the fundamental decisions for an operator to make about an Open RAN is how to split
the functions in the physical (PHY) layer of the network (which are the most demanding for
the processors) between the radio unit and the baseband (SCF238). The various split options
are defined by 3GPP. SCF (SCF225) has focused its nFAPI efforts on Split 6, which runs all the
PHY functions on the radio unit, while O-RAN Alliance, for instance, has mainly focused on
Split 7.x, which divides PHY functions between the RU and the baseband’s distributed unit DU.

12
SCF’s network FAPI (nFAPI) is an important contribution to the Open RAN platform because
it provides the most mature and deployable implementation of an open fronthaul link based on
Split 6, which is particularly well-suited to small cell, enterprise and industrial networks.

nFAPI provides a transport wrapper around SCF’s established FAPI interfaces, which
support open connections between components at the physical layer (PHY). This extends
the architecture to the MAC layer, enabling that to be virtualized, and supporting an open
fronthaul link between a small cell radio unit and a small cell distributed unit. That, in turn,
allows operators to choose these elements from different vendors if that results in the optimal
solution for their use case. The combination of FAPI and nFAPI to support Split 7 is branded as
Open6 [SCF106], [SCF159], [SCF225].

This virtualization is already specified in 4G [SCF082] and is extended to 5G by 5G FAPI. Open6

aims to establish interoperability and reduce the cost and risk to deploy small cell virtualized RANs,
by providing a uniform set of APIs, a broad innovation base and a readily deployable platform.

The choice between these splits (or others) depends on the type of network to be deployed.
Split 6, for instance, does not require high-quality fiber for every link between the RU and DU,
as Split 7.x does, because the PHY functions are in one place, and this can be more practical
and affordable in many enterprise scenarios because it reduces the need to invest in, and deploy,
optimized fiber to each radio unit, and may provide greater flexibility in the placement of cells.

13
 Figure 3. Selected approaches to functional splits in a disaggregated RAN

3GPP
3GPP gNB-CU gNB-CU
(PDCP) Option-2 (HLS) (RF, PHY, MAC, RLC)
F1

3GPP SCF
SCF S-CU S-DU S-RU
(PDCP) Option-2 (HLS) (MAC & RLC) Option-6 (LLS) (PHY & RF)
F1 5G-nFAPI

3GPP O-DU O-RAN

O-RAN O-CU O-RU
(PDCP) (Hi-PHY, MAC, (Low-PHY & RF)
Option-2 (HLS) RLC) Option-7.2x (LLS)
F1 Open-FH

Most operators plan to support at least two splits as they roll out different kinds of networks, as
demonstrated in Figure 4, which shows that the same operators will adopt each split in varying
degrees for different scenarios. For instance, about 17% of operators will adopt Split 7 in rural
environments but Split 6 in campus networks. This shows that, by 2024, in enterprise, indoor and
industrial environments, many operators will favor the x-haul flexibility of Split 6, even if they plan
to adopt a different functional split in outdoor or macro networks. This is why it is important
that there is innovation and standardization around more than one choice of split, to provide the
flexibility for operators to select the best one for their particular requirements.

Figure 4. Intention to adopt functional splits in open RAN small cell networks by 2023,
by environment (respondents were asked separately about any plans in each
environment – some selected different splits for different environments)

100 4 0
6 11 17 13
90
12 31 8
80 22 12
11 Other
% of respondents

70 14 5
Two split 07
60 16 19
19 Two split 06
50
One split 07
40 46 50
23 35 31 One split 06
30
Integrated
20
10 25 12 19
18 17
0 0 6 0
Indoor Campus Dense Rural Private
enterprise urban public industrial

The lack of a one-size-fits-all platform is unsurprising when we consider how many different
industries are interested in having cellular networks that not only improve coverage and capacity
for their buildings and campuses, but can be fully optimized for their particular processes and
services, and controlled in the same way as their other corporate networks.

14
Architecture - another dimension
of diversity in the RAN
Much discussion surrounds the need to diversify the supply base in
radio access networks (RANs). The move to more open disaggregated
RAN architectures, together with standardised, interoperable
interfaces (such as SCF’s nFAPI and O-RAN Alliance’s Open
Fronthaul) between network elements, opens up the choice of best-
in-class equipment at each network node. Vendor choice is the prime
motivation behind this industry movement. However, the benefits of
this architectural approach extend to network deployment and the
diverse challenges of real-world environments.

If you zoom in to a dense urban centre using a mapping app with

satellite imaging turned on, you’ll see a very diverse range of buildings
and environments. As you scroll north, south, east, or west you’ll
find offices, malls, stadia, parks, public transport infrastructure and
more. As you zoom out, you’ll see the variation in these environments
between dense-urban, urban, suburban and rural areas. You’ll even see
variations between environments of the same type. This rich diversity
of scenarios demands a matching diversity of RAN architecture
options to be effective in network cost and performance across the
range of challenges presented.

Small cells are at the leading edge of many of these deployment challenges
of capacity and coverage. The disaggregation of the RAN enables network
architectures that can be moulded to fit these many and varied demands
rather than force a ‘one size fits all’ approach to the market.

However, the network element cost points required cannot be met

without chips optimised to meet both the processing and transport
needs of a high performance 5G NR network. Further still, the scale
necessary for semiconductor economics cannot be met without those
chips being flexible enough to deliver the range of architecture choices
made possible by disaggregated RAN. This flexibility also supports
vendors building a range of equipment to development for a common
base of chipset and software.

Ultimately optimised, yet flexible high-performance chips are critical to

enable the diversity of RAN architectures needed in the real-world.

Doug Pulley,
Chief Solutions Architect, Picocom

15
SCF Open RAN initiatives
SCF has a long history of developing open interfaces, whether between components on a
system-on-chip; between network elements such as access points and controllers, or radio
units and basebands; or as APIs between different software functions. Enabling multivendor
interoperability, and common frameworks on which large numbers of companies can innovate,
have been at the core of its purpose ever since its foundation.

It has been very natural, therefore, for SCF’s members to take an active role in the wider
industry effort behind Open RAN. And unlike some new initiatives, there is an existing body
of work to build upon. The FAPI and nFAPI specifications are key to small cell Open RAN
and have already been through several iterations in the 4G era, and gained wide support and
adoption. 5G evolutions of both families of interfaces followed, and 5G FAPI, for instance, has
been adopted by the main providers of small cell system-on-chip products. This real world
adoption and multi-generation development greatly reduces the risk for a vendor or deployer
to support an open architecture, when compared to a brand new set of specs.

5G FAPI
5G FAPI is a particularly important contribution to small cell Open RAN because it supports
open interfaces between different components in a chipset (SCF222, SCF223, SCF224). It
provides an API (application programming interface) to support interoperability between
hardware components that implement 3GPP physical layer (PHY) functions, and the higher
layer functions that live in software (in the same integrated SC unit or on the baseband DU).

The diversification of small cell use cases and form factors and spectrum bands is attracting
new innovators into the supply chain, but all the companies that produce components for
system-on-chip (SoC) and circuit board platforms need a common framework. If an SoC
cannot be designed using components from different vendors, and then work with software
from yet more suppliers, the vendors and operators will miss out on many potential
innovations and technology combinations. The goal of the FAPI family is to provide the open
platform on which all component makers can develop and work together.

The third release of the specs, published in 2021, is important to the wider industry quest
for open, interoperable networks that span all sizes of base station and many spectrum
bands and architectures.

16
Figure 5.
The role of FAPI and nFAPI standards in integrated and disaggregated base stations

S-DU

MAC

Vendor Ext
P5 & P7
gNB/gNB-DU
Vendor Ext
L2/L3 FAPI > nFAPI

Vendor Ext
FAPI P4,P5,P7 & P19
S-RU
Vendor Ext
PHY nFAPI > FAPI

Vendor Ext
P5 & P7

PHY

5G nFAPI (Split-RAN Option 6)

Meanwhile, 5G nFAPI, first released in September 2020 (SCF225), addresses the interface
between the RU and DU, or between the physical layer and higher layer functions in the
network. Based on 3GPP Split 6, 5G nFAPI allows the RU and DU from different suppliers to
work together automatically without the operator having to invest in expensive engineering.

nFAPI enshrines all the principles of Open RAN. DU functions can be virtualized and run on
lightweight servers or edge cloud nodes, themselves based on general purpose processors; and
they can interoperate with RUs from any supplier. An open source reference implementation is
available for FAPI and can extend to nFAPI.

17
FAPI will be applied to new use cases
Radisys strongly believes that the Small Cell Forum’s FAPI interface will
be critical for SCF and O-RAN architecture-based products. The FAPI
interface will be broadly applicable in new use cases that are coming
into the market. As a leading small cell software provider, Radisys is
committed to ensuring that our software is compliant with the nFAPI/
FAPI interfaces. By implementing FAPI support in our 5G software
stack, we are making it simpler for vendors to base their SoC designs
on open platforms at every layer.

Ganesh Shenbagaraman,
Head of Integrated Products and Ecosystems, Radisys

18
 Hosted Open RAN
Not all SCF’s small cell Open RAN projects are about technical specs for the base stations.
There is broader work to understand deployers’ real-world requirements and how equipment
and business models can be devised that address these while implementing the open specs.

An example is Hosted Open RAN, a global framework that SCF proposed in 2021 to enable a
diversity of neutral host deployment models. Several large operators, such as the UK’s BT, have
pointed to neutral host or shared networks as a particularly good fit with Open RAN, but the
business model needs to be firmed-up just as urgently as the technology specs.

In enterprise scenarios, it often does not make commercial sense for more than one operator
to deploy a network, so one operator, or a neutral host, builds infrastructure that can then be
shared by the others to support all subscribers. But there are different sharing models, suited
to different environments. An established neutral host architecture such as DAS (distributed
antenna system) may work best in one scenario while others may be more effectively
supported by shared virtualized RAN platforms. However, using a mixture of sharing models
can compromise the neutral host’s business case by creating fragmentation and complexity.

Hosted Open RAN aims to provide a global framework that allows many neutral host models
to be developed within common technical and commercial specs, to bring confidence to
deployers and scale to neutral hosts and vendors.

Crown Castle embraces

Hosted Open RAN
Crown Castle embraces Hosted Open RAN as a framework that
helps facilitate decision-making and obtain cost savings. Hosted Open
RAN is attractive to neutral host operators because it lies at the
intersection of business models with shared assets and Open RAN,
bringing deployment flexibility and potential cost savings to the forefront
using a simplified methodology. Mobile operator and venue/enterprise
customers will also benefit from the adoption of the framework as it is
the result of collaboration among several global providers.

George Aguilera,
Technology Strategy Systems Architect, Crown Castle

19
SCF Member Open RAN:
Case Studies

Contents
WIA 21

Cellnex 22

Radisys 23

Picocom 24

Qualcomm 25

Druid 26

20
 Wireless
Infrastructure
Case Study Group

Wireless Infrastructure Group (WIG)

Aberdeen City Council project
Aberdeen City Council sought a solution provider that could enable high-capacity,
dedicated and contiguous coverage to over 4km-sq of the city centre with the ability to
support commercial cellular, private network functionality and O-RAN during its lifecycle.
The Council ran a competitive process and selected WIG based on our capabilities and track
record of delivering innovative solutions to meet these types of requirements.

WIG’s successful bid focussed on delivering a centralised RAN architecture at launch to

minimise aesthetic impact on the streetscape, maximise radio performance and provide
efficient, high-capacity, low-latency fronthaul for wireless service providers connecting to the
network. Dedicated dark fibre links connect node equipment (deployed on street assets) to
a central main equipment room with the ability to support a range of functional splits and
providing a more flexible migration to O-RAN.

WIG’s network in Aberdeen was the first of its kind to be deployed in Europe and has set
the blueprint for outdoor small cell networks that followed. Since its launch, WIG’s high
performance, high availability network in Aberdeen has provided a much-needed capacity boost
for those mobile network operators that have connected.

Crucially for Aberdeen, the WIG network is modular and easily scalable: as the coverage
footprint continues to grow across the city, simple upgrades to the network will allow O-RAN
architecture to be adopted enabling commercial and private wireless services to be seamlessly
activated and delivered using both MNO spectrum and Ofcom’s shared access licenses.

21
Case Study
5G Logistics – Bristol Smart Port
In partnership with The West of England Combined Authority (WECA*),
Cellnex will deploy and operate a 5G private network based on Open RAN
architecture and coordinate the development 5G innovative services for Bristol Port and
Gravity Smart Campus.

The project, awarded by the UK Government (DCMS 5G Programme) will demonstrate how Open RAN
5G private networks can offer efficiency and productivity improvements to the logistics ecosystem by:

• Improving port estate management and the efficiency of logistics arrival/handling/despatch, including
allowing real-time location tracking of individual items,

• Allowing improvements to road traffic management, including traceability and continuous geolocation,
and monitoring goods and materials to coordinate movements between Freeport and linked zones, and

• Replacing low value, manually intensive processes, with 5G-enabled autonomous smart systems.

5G logistics project will establish a blueprint for smart and dynamic port environments. It will focus on
security, traceability, and real-time monitoring within and across extendable virtual boundaries and between
public and private networks.

5G small cells will be strategically deployed to cover the different test zones: Freeport locations for
tracking and drones use cases, Gravity and St Andrews Road for Smart Junction Use Case.

22
Case Study
Release 16 Compliant
5G NR Software
Radisys’ Release 16 compliant 5G NR protocol Trials and test that have
software supports low latency use cases for 5G
private networks. Radisys’ Connect RAN 5G been completed or are underway
solution, based on specifications from 3GPP,
Radisys has several trials underway with partners
the Small Cell Forum, and the O-RAN Alliance,
that leverage its Connect RAN Small Cell Software.
enables an elastic and agile 5G network tailored
We are working with a variety of ODMs to deploy
to meet the unique requirements of diverse 5G
our software on their chipsets and platforms to
applications with minimized CAPEX, improved
enable low power, indoor small cells. These small
time-to-market, and shortened time-to-revenue.
cell solutions are being used in global trials for
The new Release 16 compliant 5G NR protocol
a variety of different use cases such as indoor
software delivers enhancements in functionality,
coverage and public safety. With our mature
capacity, coverage, latency, mobility, reliability, and
software, we have been successful in enabling
ease of deployment.
nFAPI/FAPI-based small cells that operate in sub-6
• Release 16 compliant protocol software unlocks GHz and mmWave. Our partners include:
the full potential of 5G NR, enabling operators
• Qualcomm: Radisys’ 5G Small Cell Software
to monetize their networks by supporting new
has been integrated on the Qualcomm FSM
use cases for industry verticals. It delivers new
100xx 5G Platform, delivering a complete, pre-
features, including URLLC (Ultra Reliable Low
integrated, pre-tested, and pre-validated 5G
Latency Communications), TSN (Time Sensitive
small cell solution for both mmWave and sub-6
Network), NR Unlicensed, 5G Location and
GHz bands for FDD and TDD deployments.
Positioning Services and UE Radio Capability
Radisys 5G small cell solution has been
Signaling Optimization.
benchmarked in operator lab trials delivering
• The upgraded protocol software, with new Gigabit throughput eMBB services supporting
enhancements for beam management for 5G multi-cell and carrier aggregation features.
NR sub-6 GHz and mmWave frequency bands,
• Picocom: Radisys and Picocom completed
MU-MIMO, New Radio Dual Connectivity (NR-
interoperability testing of SCF FAPI interface
DC), Integrated Access and Backhaul (IAB), and
using Picocom’s PHY Software Integration Kit
Dynamic Spectrum Sharing (DSS), enables robust
with Radisys’ 5G Small Cell Software. This end-
and efficient network capacity and coverage.
to-end testing demonstrated real time video
• Radisys’ Connect RAN 5G Software won streaming on Picocom PC802 platform.
the SCF Small Cell Award for Outstanding
• NXP: Radisys 5G Small Cell Software has been
Contribution to Small Cell Open RAN
integrated on NXP 2160 platform with PHY
Platforms. The software is available pre-
layer from partners to provide FAPI based small
integrated and pre-validated on multiple
cells and CU/DU solutions. We demonstrated
industry-leading platforms to accelerate global
this solution in ARM 5G RAN Summit 2021.
5G deployments.

23
Case Study
ORANIC Open RAN PHY and NIC board
The PC802 ‘ORANIC’ is an Open RAN PHY and NIC board and contributes
to the diversity of RAN architectures. It is an SCF Small Cell Awards 2021 winner
for innovation in easing small cell deployments. It targets O-DU applications in varying
deployment scenarios.

ORANIC is a PCIe board designed for 5G/4G O-RAN O-DUs in conjunction with
L2 software running on a server. ORANIC and its carrier grade PHY software utilises
industry leading Open RAN standards:

• SCF FAPI interface for layer 2 messages

• O-RAN Alliance Open Fronthaul (OFH) interface

The card handles all upper PHY and OFH functionality for 4G and 5G small cells,
significantly reducing the load on servers. ORANIC integrates four Picocom PC802
chips, which provide for four 25 Gigabit Ethernet SFP connectors. Each board can
support up 32 antennas over multiple RUs over the O-RAN Open Fronthaul interface,
supporting O-RU configurations including 16 2T2R or 8 4T4R.

24
Case Study
Qualcomm Technologies and T&W demo
an Open RAN white box based on Split 6
Qualcomm Technologies, Inc. worked with T&W to demonstrate a unique white box
solution for small cell Open RAN. This combines elements of the Small Cell Forum and
O-RAN specifications to maximize flexibility and cost efficiency, and to make roll-out
as quick and simple as possible, with significant use of existing IT infrastructure.

The demonstration showed a disaggregated small cell based on functional Split 6

and SCF’s nFAPI fronthaul interface. This architecture has the advantage of relatively
relaxed requirements for fronthaul capacity and latency, because the most demanding
Layer 1 network functions are left on the radio unit (RU), reducing the burden on the
distributed unit (DU) and its fronthaul connection to the RU.

This enables deployers to use off-the-shelf Ethernet connectivity, which may already be
installed on a site, perhaps for backhauling WiFi access points, and off-the-shelf servers
to support the DU. In turn, this saves on cost and complexity by reusing existing
infrastructure and avoiding the need for expensive hardware acceleration to support
Layer 1 functions on the DU.

The architecture does require relatively sophisticated RUs, though their cost is offset
by the efficiencies of using COTS IT infrastructure. The approach is well suited to
large deployments to support challenging use cases with high performance radio
requirements. These might include multi-operator networks within a large venue, or
large-scale industrial private RANs that support applications such as robotics or real
time machine control.

In such environments, large numbers of RUs need to be managed in harmony, often

supporting technologies such as coordinated multipoint (CoMP). To optimize this,
Qualcomm Technologies and T&W used elements of the O-RAN platform, notably its
synchronization specs and its management plane design. These can coordinate the RUs
in a highly centralized way. They can also manage RUs from multiple suppliers, based on
the O-RAN RU M-plane, a subset of the full M-plane architecture.

25
Case Study
SmartRAN Open Network
Interoperability Centre (SONIC)
The SmartRAN Open Network Interoperability Druid Software Raemis 5GC
Centre (SONIC) is a joint programme SA standalone core platform was
between Digital Catapult and Ofcom (UK) to chosen for the SONIC 5G ORAN test beds
test interoperability and integration of open for several reasons, for example:
networking solutions, vendors and system
integrators. Funded by the UK government and • The Raemis platform is 3GPP compliant
Ofcom, SONIC aims to help achieve a vision of with unique features designed specifically for
a more open, competitive, and diverse telecoms enterprise critical communications.
supply chain. This project is a great example of
the collaboration between Core, CU, DU and RU • The platform is differentiated by its ability to easily
from the many partners in the ecosystem. prioritize and “slice” devices while providing them
with a guaranteed quality of service.
Digital Catapult existing testbeds in London and
Brighton are now operational, coupling laboratory • Interoperability with any small cell RAN vendor,
capabilities with real world indoor and outdoor with 25x 5G small cells tested with Raemis to
deployments. SONIC begins with Open RAN and date, gives partners greater options to choose
focuses on open software-centric networking. In from indoor or outdoor specialists.
addition to telecoms solution providers of all sizes,
• Raemis scales up as easily as it scales down,
SONIC will supply insights to network providers
enabling the setup of small or very large
– including national operators and smaller private
private networks.
and neutral networks.
• Raemis can be deployed on edge hardware
SONIC will foster an ecosystem in the UK,
devices or in high dependency cloud native
of large and small telecoms suppliers, with
environments. This flexibility is key for certain
multiple providers for each element in the
highly resilient mobile use cases or customers
technology stack. Large businesses can explore
who do not want their data on the public cloud.
future deployments such as private and virtual
networks, while smaller companies can test the • Built on an open REST API, Raemis provides
integration potential of their early-stage products. easy integration to enterprise critical applications
Providing a neutral environment for testing and and makes private networks simple to use and
demonstration will surface the opportunities and manage with a user-friendly dashboard.
challenges of interoperability.

26
Conclusions
The various Open RAN initiatives and early deployments will lay the
foundations for a body of technology and knowledge that will increase
confidence and accelerate adoption as the decade progresses. Much of
the momentum will come from enterprise environments, as Figure 6
shows. This summarizes our moderate-case forecast for deployment
of small cell Open RAN (by radio units) in the first half of the 2020s.
By 2026, enterprise deployments will be running at twice the volume
of those in public networks, where lower numbers of small cells will
be needed than in dense indoor and B2B locations.

However, smart city projects and other developments will start to drive
densification in public networks and small cell Open RAN can help to
improve the challenging economics of that process. Between 2020 and
2026, then, enterprise Open RAN small cells will grow at a CAGR of
72%, while those in public networks will grow by 114% CAGR.

This rate of growth will be driven by open architectures that enable many
deployers, vendors and service providers to participate in the market in
every area, from components to applications. That will lower the cost
of development and deployment, so it will be cost-effective to create
solutions for specific industries or use cases, improving the business case
for all stakeholders, as our case studies demonstrate. Further, open small
cell networks will work with other key trends such as spectrum sharing
and infrastructure sharing, to accelerate the adoption of innovative new
operational models such as neutral host.

Figure 6.
Deployment of small cell Open RAN (by radio units)

3,500

3,000 Enterprise Public

2,500
,000 RU

2,000

1,500

1,000

500

0
2020 2021 2022 2023 2024 2025 2026

Year

27
Join Small Cell Forum today!
SCF is a global membership organization committed to supporting
agile, low-cost mobile infrastructure through small cells. Our members
include MNOs, neutral hosts, towercos, equipment and technology
vendors, and service providers.

Find out more, email info@[Link]

[Link] | April 2022
28