LTE Parameter Tuning

BoR (18) 23
Study on Implications
of 5G Deployment on
Future Business Models
No BEREC/2017/02/NP3
A report by DotEcon Ltd and

Axon Partners Group
14 March 2018
Contents
Contents
1 Introduction .............................................................................................................................................. 1
2 Background and context: the development of 5G ..................................................................... 2
2.1 Defining 5G ...................................................................................................................................... 2
2.2 Global trends in information and communications technologies .............................. 4
2.3 The roadmap towards 5G ........................................................................................................... 7
2.4 Regulatory environment.......................................................................................................... 11
3 Overview of key 5G technologies .................................................................................................. 20
3.1 Technical capabilities that differentiate 5G ...................................................................... 21
3.2 Enabling 5G access using a combination of technologies .......................................... 24
3.3 A single network to serve multiple demands .................................................................. 26
4 Use cases for 5G .................................................................................................................................... 28
4.1 Automotive ................................................................................................................................... 30
4.2 Media and entertainment ....................................................................................................... 35
4.3 Manufacturing ............................................................................................................................. 39
4.4 Logistics ......................................................................................................................................... 42
4.5 Agriculture .................................................................................................................................... 45
4.6 Energy and utilities .................................................................................................................... 49
4.7 Healthcare ..................................................................................................................................... 53
5 Business models and value chains ................................................................................................. 58
5.1 How might 5G enable new business models? ................................................................. 59
6 Drivers and obstacles to 5G .............................................................................................................. 69
6.1 Enhanced mobile broadband as the initial driver .......................................................... 69
6.2 Identifying new revenue sources from additional capabilities of 5G ...................... 76
7 Regulatory options .............................................................................................................................. 84
7.1 Challenges arising from small cells ...................................................................................... 84
7.2 Access to spectrum .................................................................................................................... 88
7.3 Coverage issues........................................................................................................................... 92
7.4 Competitive issues ..................................................................................................................... 94
7.5 Backhaul for 5G cell sites ......................................................................................................... 96
7.6 Edge computing ......................................................................................................................... 96
7.7 Net neutrality ............................................................................................................................... 97
i
Contents
7.8 Sectoral regulation and quality of service ......................................................................... 99

8 Conclusions .......................................................................................................................................... 100
Annex A Glossary ................................................................................................................................... 102
Annex B Automotive case study ...................................................................................................... 105
Annex C Media and entertainment case studies ....................................................................... 108
Annex D Manufacturing case study ................................................................................................ 113
ii
Tables & Figures
Tables & Figures

Table 1: Role of 5G in global telecom trends .................................................................................... 5
Table 2: Potential use cases in the automotive industry ........................................................... 32
Table 3: Potential use cases in the media and entertainment industry ............................... 37
Table 4: Potential use cases in the manufacturing industry .................................................... 40
Table 5: 5G Opportunities via convergence of technology and logistics trends ............. 43
Table 6: Potential use cases in the logistics industry .................................................................. 44
Table 7: Potential use cases in the agriculture industry ............................................................ 47
Table 8: Potential use cases in the health sector .......................................................................... 54
Table 9: Examples from case studies ................................................................................................ 78
Figure 1: Evolution of cell data rates under different radio access technologies ............... 2
Figure 2: Expected roadmap towards 5G .......................................................................................... 8
Figure 3: The main families of usage scenarios and applications 5G might support ..... 20
Figure 4: Technical capabilities of IMT-2020 .................................................................................. 21
Figure 5: Enhancement of key capabilities from IMT-Advanced to IMT-2020................... 22
Figure 6: Enhancement of key capabilities from IMT-Advanced to IMT-2020................... 22
Figure 7: Key technologies to enable 5G ........................................................................................ 25
Figure 8: 5G virtual end-to-end networks tailored to serve application requirements . 26
Figure 9: Overview of proposed verticals and illustrative case studies ............................... 28
Figure 10: Value chain with service differentiation under 5G ................................................. 60
Figure 11: Business models to support embedded connectivity ........................................... 62
Figure 12: The value chain considering cell densification and access to small sites ...... 65
Figure 13: Convergence of PPDR-MNO-FWA provision ............................................................. 66
The content of this report does not reflect the official opinion of the Body of European
Regulators for Electronic Communication. Responsibility for the information and views
expressed therein lies entirely with the authors.
i
Executive summary
Executive summary
5G is evolutionary It is commonly supposed that 5G will lead to a step-change in the
rather than capability of mobile networks, opening up possibilities for
revolutionary
innovative new services, such as the Internet of Things (IoT),
connected vehicles and augmented reality (AR). Whilst 5G may
eventually be transformative for some sectors, in the short to
medium term these developments are more likely to be
evolutionary rather than revolutionary. The speed of roll-out and
adoption of 5G will depend on complementary investments being
made in network infrastructure as well as new services, applications
and products using 5G.
What 5G might bring
Network slicing will 5G brings a number of enhancements over 4G, including high
allow differentiated speeds, low latencies, enhanced reliability, lower power
services
consumption and greater terminal device densities. Perhaps most
important, 5G offers new network management possibilities that
could enable a single physical network to support a number of
virtual networks with different performance characteristics.
This network slicing creates, for the first time, the possibility of
tailoring mobile data services to the particular characteristics of
specific users. For example, a dense IoT sensor network might
prioritise low power consumption of terminals over connection
speed; at the same time, a separate network slice on the same
infrastructure could deliver high-speed mobile broadband.
New business models
Services can be This new ability for differentiation of services without having to
targeted to the needs build different physical networks raises the possibility of services
of verticals and
targeted at particular economic or industrial sectors – so called
specific user groups
‘verticals’ – as well as at specific user groups. Therefore, 5G has the
potential to change business models for network operators relative
to the current marketplace, where network operators have offered
largely standardised services and differentiation has been limited to
pricing plans.
There may be new This also opens up potentially new roles for intermediaries in the
intermediaries value chain, positioned downstream of network operators, offering
to bundle and repackage connectivity for particular industries. For
example, aggregators might put together trans-national
connectivity packages for particular industries. In general, there are
likely to be opportunities for orchestration of different networks,
tying them together to create connectivity services for specific
verticals.
ii
Executive summary
Upstream entrant of There may also be opportunities for new players upstream of
wholesale cell traditional mobile networks. For example, 5G is likely to require
densification services
significant densification of current networks at high frequencies to
meet bandwidth requirements. This may create new opportunities
for independent operators to acquire sites in dense urban areas and
indoor public spaces and deploy 5G infrastructure, offering
wholesale services providing patches of connectivity to 5G
operators. Given availability of appropriate spectrum, traditional
network operators could also be bypassed by entrants focussed on
providing connectivity for specific industries (for example, within
factories or warehouses).
Entry should not be Spectrum award designers need to be aware of these possibilities
prevented by for upstream entry, rather than assume that the only buyers of
spectrum award
spectrum will be incumbent MNOs. Spectrum should be packaged
design
to allow entry, for example by allowing bidders to assemble smaller
blocks to give flexibility over the amount of spectrum acquired.
Drivers of 5G roll-out
eMBB, is the initial Despite these transformative possibilities of 5G in the long-run,
driver of 5G there is currently broad consensus amongst stakeholders that initial
deployments of 5G will be driven by enhanced mobile broadband
(eMBB). Operators will be incentivised to deploy 5G by lower unit
costs of network capacity and the need to maintain competitive
service quality in the face of continued data growth.
eMBB may generate It is possible that eMBB alone will not create significant additional
limited additional revenue from 5G for network operators. Experience of previous
revenue
migrations from one mobile technology to the next suggests
consumers pay a broadly similar amount even though data speeds
have increased and bundled data and call allowances have grown
due to falling unit network costs.
Prospects for additional revenue sources
Long-run incentives In the longer run, 5G roll-out incentives depend on incremental
depend on revenues from new services. In turn, incentives to develop and
additional revenues
market novel services dependent on 5G will require sufficient 5G
from new services
coverage, which creates a coordination issue.
Developers of Any commercial developer of a new service or product benefiting
applications and from 5G connectivity will want to reduce its exposure to the risk of
services will mitigate
5G roll-out being slow or geographically limited. We have found
risks by having fall-
back connectivity that developers tend to follow diversified approaches to
options connectivity, looking to develop services that will fall back smoothly
onto slower or less capable mobile networks or which might even
use quite different technologies. For example:
• Agricultural applications of dense sensor networks are
unlikely to wait until 5G becomes ubiquitous in rural areas.
Rather, developers wanting to get new products and
services to market will adopt other interim technologies
that are readily available. For example Long Term Evolution
for Machines (LTE-M) or NarrowBand IoT (NB-IoT), or low-
iii
Executive summary
power wide-area technologies such as LoRa in a hub and

spoke arrangement with centralised backhaul, possibly
even using satellite.
• So-called ‘vehicle to everything’ (V2X) communications
have a development path through 4G and are not
dependent on 5G availability. Connected vehicles will need
to make use of a wide range of connectivity strategies,
include fall back to lower specification networks and peer-
to-peer networking. We understand that car manufacturers
are keeping all these options open, rather than relying on
5G availability. In the short term, 4G cellular vehicle
connectivity will be used to deliver infotainment.
• A leading national broadcaster indicated that because of its
public remit to make services available widely, it was
concerned about developing new services using 5G
capabilities if this would lead to a digital divide with those
services only being available in urban areas and would need
to have fall-back options.
Lack of a killer application
Incremental For some applications, 5G will have a convenience yield. However,
revenues from 5G are willingness to pay for 5G services in the short-to-medium term
limited by substitute
might also be limited by the availability of alternatives, including 4G
technologies…
and faster RLANs for example.
…but there may be It is likely that incremental revenues from 5G will derive first from
many niche niches where 5G provides capabilities that are not readily or cost
applications where
effectively met by alternatives. Examples are services requiring very
the 5G is needed, just
no killer application low latency over a wireless connection crucial for real-time control
and safety critical applications, or those where high reliability and
dedicated capacity may be very important for critical machine-to-
machine communications, especially where the number of devices
is massive.
Much of the discussion of 5G to date suggests that there are certain
sectors where 5G will be important – or even critical – such as
transport, medical applications and, more generally, IoT. However,
we disagree that there are certain sectors that, by themselves, can
provide significant additional new revenues within a reasonable
timeframe to drive 5G roll-out. Rather, 5G is much more likely to
have a large number of niche applications across many sectors,
meeting particular needs that are not already well met by 4G or by
alternative technologies.
The impact of 5G may in the long run be pervasive and much
broader than the usual case studies suggest; at the same time, there
is a challenge in identifying these various likely demands for 5G and
creating services and pricing arrangements targeted at those uses.
iv
Executive summary
Pricing structures and efficient take-up

Some pricing Take-up may depend on providers establishing a suitable charging
structures could lead structure that is attractive relative to the alternatives. There is a
to inefficient
danger that over-reliance on per-connection charging models could
outcomes
distort users’ choices towards less efficient alternative technologies.
Pricing structures need to balance benefits to new users and cost to
the network provider; this may require certain risk-sharing
arrangements between network operators and verticals.
Private networking
Private 5G networks We see considerable scope of private 5G networks within the
should encouraged confines of factories and warehouses. This is pro-competitive and
will help drive broader 5G roll-out. It is important that spectrum
allocation models take this possibility into account. Unlicensed
spectrum may not be appropriate for these applications due to their
reliability and security needs. A secondary licensing model could
work well at higher frequencies with limited propagation.
There may be need for some degree of coordination about which
bands should be used for private 5G networks and the licensing
model for spectrum across Member States. This is a matter for
possible guidance from BEREC.
Consequences of small cells
Site owners may gain A particular issue is the possibility of site owners gaining increased
power due to small power within the overall value chain due to the difficulties of
cells
finding appropriate sites for small cells, especially in dense urban
areas and ‘quasi-public’ spaces such as stadia and shopping centres
that are privately controlled.
Owners of pivotal sites may have a degree of market power and
might even have an incentive to limit access to network operators
to extract more rent from those operators that are present. Any
distortion of downstream competition in mobile services can be
removed with obligations on any network operator present to make
a corresponding wholesale service available (in effect a neutral
host).
Governments can This measure does not entirely prevent rent extraction by pivotal
take action to site owners. However, their power can be reduced by increasing the
increase the supply
supply of sites. Planning constraints should not be excessively tight.
of sites
Public sector bodies may control appropriate site for small cells in
urban areas (e.g. street furniture or roofs of social housing); there is
a danger that in seeking to maximise their commercial returns,
these bodies might have a degree of market power.
It is possible that fixed operators might have certain advantages
over pure mobile operators in deploying small cells, for example
bundled in consumer routers to support FTTP services.
v
Executive summary
Availability of backhaul
Rural 5G deployments need backhaul and, in some cases, the
necessary fibre infrastructure may be lacking. It may be possible to
piggy-back on interventions aimed at encouraging high-speed
broadband in rural areas. Physical infrastructure access may help to
some extent, but this may not be particularly effective in
encouraging fibre in rural areas if physical infrastructure is lacking.
Infrastructure sharing
5G is likely to lead to much more infrastructure sharing, due to the
use of small cells. Sometimes there may be physical limits on how
many distinct networks can be accommodated at a site.
Regulators need to be alert to excessive concentration at the
network level due to infrastructure sharing. Charging models for
shared infrastructure should ideally be based on capacity-based
charges to ensure that there are no retail competitive effects.
However, we question how effective this can be in protecting
competition, when the agility of 5G networks might allow rapid
reallocation of capacity between sharers.
Edge computing
It is possible that applications such as augmented reality might use
edge computing for low latency applications. Edge computing can
only be procured from 5G network operators. Open standards for
portability of edge computing applications may be important to
ensure that consumers are not locked in to one network operator.
vi
Introduction
1 Introduction
The Body of European Regulators for Electronic Communications
(BEREC) has commissioned DotEcon Ltd (DotEcon) and Axon
Partners Group Consulting SL (Axon) to conduct a study on
implications of 5G deployment on future business models.
This study aims to help BEREC improve its understanding of 5G in
terms of the key technologies and services that may emerge, and
the obstacles and enablers for success, with a focus on the near
term, specifically the time period from 2018-2022. In particular,
BEREC would like to understand the implications of 5G for market
structure, market dynamics and competition in the
telecommunications market and other related markets.
In line with these objectives, our report is divided into six sections:
Background and context to the development of 5G - This section
summarises the main European 5G initiatives and regulatory
measures.
Key 5G technologies - This section provides an overview of the key
aspects of 5G technology and their implications for services,
including developments in the radio access network (RAN) and
underlying network infrastructure. We consider the progressive
‘evolution’ from 4G/LTE-A, but also how 5G might allow for more
fundamental changes than the shift from 3G to 4G in terms of
enabling services that were not previously possible.
Identifying potential use cases of 5G – We consider the main
‘vertical’ industries typically considered to be potential beneficiaries
of 5G and consider possible use cases within these sectors. We ask
which of these use cases are likely to be important in providing
incentives for 5G roll-out and identify any particular issues or
barriers to 5G adoption in those sectors.
Potential implications of 5G deployment on business models
and value chains - We consider the possible impact of 5G on
business models and value chains within telecoms and vertical
sectors and consider any new opportunities that may arise,
including the potential for an increasing role for intermediaries. This
may have implications for both the telecoms and adjacent markets
in terms of market structure, competitive impacts and charging
models to monetise 5G network investments.
Drivers and obstacles – In this section we consider the main
drivers and obstacles to deployment of 5G services.
Regulatory options - Having identified the drivers and inhibitors,
we will consider the key questions BEREC and NRAs should consider
when assessing how the regulatory environment can be set to
facilitate successful deployment of 5G services.
1
Background and context: the development of 5G
2 Background and context: the

development of 5G
2.1 Defining 5G
The next generation 5G is the term used to describe the next (and fifth) generation of
of wireless networks wireless networks, beyond current 4G LTE networks. Although a
final standard for 5G is yet to be formally designed,1 5G networks
are expected to build on, and smoothly integrate with, the legacy of
previous generations of wireless network. In the first instance, 5G
represents an evolution of existing radio access technologies.
From 2G to 4G, each radio access technology generation over the
last 25 years has focused on improving the speed and efficiency of
wireless networks to enhance mobile services.
Figure 1: Evolution of cell data rates under different radio access technologies
Voice
2G 64 K bps
Voice and D ata
3G 2 Mbps
D ata
4G 1 Gbps
D ata and m ore…
5G Up to 20 Gbps
Source: DotEcon and Axon
Each transition has been driven by the greater capabilities of each

generation over its predecessor. 3G enabled mobile internet and
data connectivity at a much more advanced level than 2G.2 4G was
able to serve the massive increase in mobile data traffic, something
3G was not capable of.
With more people using mobile broadband as services and
applications expand, more devices of various types connecting to
mobile networks and continued growth in data, further
enhancements will eventually be required. Operators are looking
1
Although, 3GPP Release 15 has defined the 5G system architecture. See 3GPP
Release 15: www.3gpp.org
2
The emergence of (limited) data connectivity and mobile internet predated “3G”
in the form of General Packet Radio Service (GPRS) (Release 97).
2
ahead to identify ways their networks can be readied to meet future

capacity and performance requirements.
5G will support The deployment of 5G networks is seen as important in fulfilling
expected mobile expected mobile data traffic growth. However, 5G differs from
data growth
previous generations in important ways. Improvements to existing
mobile broadband services to provide enhanced mobile broadband
(eMBB) is just one use-case for 5G. In the longer run, 5G is expected
provide tailored connectivity to meet the particular demands of
different user groups, including particular industries (so-called
‘verticals’). Verticals may need data connectivity both to use
internally for their own operations and to embed within the
products and services they supply to their end customers.
5G will allow for the 5G will create opportunities for network operators to tap new
emergence of new sources of revenue by developing infrastructure that can deliver a
services.
range of innovative services for enterprises, including IoT (Internet
of Things) applications and deep integration of connectivity (e.g.
transport telematics). For the first time, mobile networks will be
designed to address the varying needs of different industries and
the rise of connected devices (so-called massive machine-type
communications – mMTC). As more devices and objects need to be
securely, automatically and remotely connected and monitored –
allowing systems, machines and infrastructures to run with end-to-
end machine-to-machine communication – a large number of
sectors may become increasingly reliant on wireless network
solutions.
5G brings network 5G is projected to co-exist with the 4G networks, but to deliver
performance improved connectivity through high speed, reliable and secure
enhancements
communications to approximately one-third of the global
population by 2025, five years after its launch3.
5G is able to bring enhanced capabilities (a topic discussed further
in Section 3), including lower latency, high resource efficiency,
decreased energy consumption, and enhanced security. This can
drive access to a broad range of applications and services, as well as
opening opportunities for new business models within the
telecommunications sector and key verticals.
The software and IT industry in Europe stands to evolve because the
5G network services will rely on software and virtualisation, creating
an expanded ecosystem. This may create options to develop new
partnerships and synergies across industries.
5G brings agility in 5G will allow for a shift from networks designed and built from the
network outset for specific performance characteristics to agile networks
characteristics
that can be programmatically assembled and configured for specific
use cases. In effect, a common network infrastructure can carry
3
GSMA Intelligence, Global Mobile Trends, 2017.
3
multiple ‘virtual’ networks with differing performance

characteristics aimed at different types of users, with this being
readily reconfigurable without needing further physical investment.
2.2 Global trends in information and

communications technologies
The requirements for 5G ultimately derive from established, global
ICT trends, as more industries look to ICT both to improve their
productivity and to develop attractive new services and products
for their customers. This creates demand for faster connections,
increased portability, higher quality experiences and increased
security from mobile data.
Table 1 below shows some of the emerging trends in the ICT sector,
discussed in more detail below. These are likely to be significant
drivers of growth, innovation and disruption across many industries;
5G may have an important role in supporting them.
4
Table 1: Role of 5G in global telecom trends
Major trend Key role of 5G
Connectivity: growth in devices, content and data • Handle growth in a cost-effective, high-speed and
traffic energy-efficient manner.
Proliferation of value-added services and over- • Support evolution of richer content types such as 4K,
the-top (OTT) players: operators adapting to 8K, VR, AR, 360° videos etc.
changing times
• Enable telecoms operators to compete with OTT
players through partnerships and quality offerings.
Fixed Wireless Access (FWA) technology: the great • Utilise much higher frequency bands than current 4G
wireless migration networks.
• Solve the last mile problem4.
Digitisation and advanced analytics: achieving • Digitise and overhaul business models of telecom and
maximum value from each customer adjacent industries.
• Provide data analytics to accurately segment and
generate maximum value from each customer.
Internet of Things (IoT): explosion of connected • Connecting everything from smart home devices to
devices self-driving cars and robots.
• Improve the latency capabilities of current 4G LTE
systems, leading to increased efficiency.
Network upgrades and security: need for security • Ensure quality, security and safety.
by design • Transform security to be the new area of competitive
advantage for operators.
Consolidation, bundling of services and M&A: • Increase consolidation and M&A activities amongst
attractive vehicle for entering new markets operators to gain returns on investments in 5G
networks.
• Increase network sharing to achieve densification.
Connectivity: There has been a massive increase in demand for

connectivity driven by smartphones (and other connected devices)
becoming a channel for consuming content such as music and
video, running a large number of applications including payments
and e-commerce, and remotely controlling devices of various types.
In the longer run, it is unrealistic to expect sustained - or even
accelerating - rates of data growth to come from consumer devices
alone. Long-run growth in data volumes may also come from new
services, including connected devices communicating directly
machine to machine. 5G is expected to have a role in meeting this
growing demand by both delivering enhanced mobile broadband
and opening up new opportunities serving industries.
4
Last mile is the portion of the network that reaches the user´s premises. Bulk of
the costs and most of the complexity involved in fixed access deployments are
associated with the last mile.
5
Proliferation of value-added services and over-the-top (OTT)

players: OTT (over-the-top) players have continued to make gains
in core communication services such as messaging and voice. 5G
could help support OTT services based on even richer content
types.
Fixed Wireless Access (FWA) technology: Fixed Wireless Access
(FWA) is an established means of providing internet access to
homes using wireless network technology rather than fixed lines.
FWA might become an alternative to fixed broadband access due to
increased spectrum availability and technological progress in the
4G and 5G arena. Although FWA can be achieved through current
mobile technologies, it cannot offer the download speeds or
latency levels that might be demanded by the users in future nor is
it currently a cost-effective option in fixed locations. 5G-based FWA
might enable more robust services with improved capacity and at a
lower cost. FWA based on 5G would also be a stepping stone to full
5G mobility, possibly contributing to a smoother and swifter
transition from 4G to 5G for users.
Digitisation and advanced analytics: Digitising business
operations can make companies more productive and improve
customer satisfaction, all of which lead to increased revenues.
Advanced analytics can allow businesses to segment customers and
allowing them to predict customer behaviour, deriving more value
from each customer. 5G´s ability to support data collection and
distributed processing through edge computing might assist
certain aspects of data analytics and digitisation.
Internet of Things (IoT): Significant growth in the number of
connected IoT devices is expected, with a CAGR of 21% per annum
until 20225, driven by a growing range of use cases and the falling
cost of devices. Mobile network operators are well placed to offer
connectivity for emerging IoT applications due to their existing
coverage. Although alternative technologies may be sufficient at
present, including improvements being made to 4G services to
support such applications, 5G might achieve the reliability, latency,
scalability, security and ubiquitous mobility that could support
proliferation of a mass-scale IoT environment.
Network upgrades and security: The demand for improved
customer experiences and cost-efficient network operations are
currently driving network upgrades. Further improvement in
network coverage may be important for serving some verticals.
There will also be increasing demands for network security and
integrity from Governments, businesses and consumers. 5G
5
Ericsson, “Mobility Report”, 2017. Available at:
https://www.ericsson.com/assets/local/mobility-report/documents/2017/ericsson-
mobility-report-june-2017.pdf
6
networks will be aiming to serve not only individual customers, but

also other industries in providing diverse services. Through scalable
identity management, distributed authentication and network slice
security amongst others, 5G will be capable of building a more
efficient and effective security solutions.
Consolidation, bundling of services, and M&A: Acquisitions are
an attractive vehicle for telecom players to create product offerings
and services beyond their traditional capabilities and for entering
new markets and businesses. We have already seen operators
expanding the range of their offerings through quadruple play of
services of telephone, television, broadband and wireless. Some
transactions are also targeted at entering adjacent industries. The
level of cost and capabilities required to achieve scale in 5G might
become a further factor in M&A plans among network operators,
communications companies, and mobile and fixed operators.
2.3 The roadmap towards 5G

The industry has generally accepted that the start of commercial
deployment of 5G systems is expected to take place around 2020,
following research and development and the conclusion of
standardisation and regulatory initiatives.
It is perhaps too early for (European) operators to commit to
network rollouts, but many are predicting commercial availability of
5G around 2020 to 2025. The exploratory phase to understand
detailed requirements for 5G future systems and to identify the
most promising technical and technological options has already
started. Ambitious 5G trials and pilots are currently being
implemented in various parts of the world.
Figure 2 below illustrates the major expected milestones on the
journey to 5G availability with respect to standardisation, spectrum
release and global initiatives and events.
7
Figure 2: Expected roadmap towards 5G
5G Ex perim ents
D eployment 5G Trials
of 5G
5G D eploym ent and Com m ercializ ation
Vision, Workshop Evaluation

ITU
Proposals IMT-2020 IMT-2020 S pecifications
S tandardisation
Release 15 – Release 16 – Phase2 Phase 1
Phase 2 Deployment
Phase1 Standards Standards Deployment
3GPP
5G N R Rel- 15 Full
N on Stand (Incl Stand
Alone Alone)
S pectrum
WRC 2019 WRC 2023
release
• Uninterrupted 5G coverage in all urban areas and

5G Targets • At least one major city per member state to be 5G enabled
major terrestrial paths by 2025
5G Private Trials, 5G Platforms, 5G for Euro 2020,

Pan EU Trials 5G Vertical Pilots, 5G Trials Cities
Global
initiatives S pecification
And events 5G PPP Development/Tests
Experimentation/Pilots
Events
2017- 2018 2019 2020 2021 2022 2023
Standardisation Relevant standard bodies, both regional and global, have set out
supported by ITU and timetables for the development of 5G. For example, ITU and 3rd
3GPP
Generation Partnership Project (3GPP) are focused on studying
technological requirements for 5G and releasing the standards by
2020. In 2012, ITU-R6 embarked on a programme to develop IMT for
2020 and beyond, setting the stage for 5G research activities
currently emerging around the world. IMT-2020 is set to be the
global communication network for the coming decades and is on
track to be in place by 2020.
The core of 5G standardisation related to mobile technologies will
happen in the context of 3GPP standardisation.7 This foresees two
phases of 5G deployments:
• 5G New Radio (NR)8 features set out in 3GPP Release 15
will form the first phase of 5G deployments;
6
ITU Radiocommunication sector (ITU-R) is one of the three sectors of the ITU and
is responsible for radio communication.
7
3GPP Release 15 and Release 16.
8
5G NR is the wireless standard that will become the foundation for the next
generation of mobile networks.
8
• Full compliance with ITU´s IMT-2020 requirements is

anticipated with the completion of 3GPP Release 16 in
2019, forming the second phase of 3GPP´s 5G deployments.
In December 2017, 3GPP announced the completion of Release 15
NSA 5G NR specification, which introduced a relevant milestone to
complete the first implementable specification for Non-Standalone
5G NR operation. The Non-Standalone mode revolves around the
enhanced Mobile Broadband (eMBB) use case and will use existing
LTE radio and an evolved packet core network as an anchor for
mobility management and coverage while adding a new 5G NR
radio access carrier. This is the configuration that will be the target
of early 2019 deployments. This mode will be a cornerstone of 5G in
the early years of adoption as mobile carriers introduce network
compatibility. The completion of this specification gives the
industry the green light to accelerate design and implementation of
equipment adhering to the standard. The completed specifications
have support for low-frequency (600 MHz, 700 MHz bands), mid-
range (3.6 GHz9 band), and high frequency (50 GHz band) spectrum.
The technical specifications for Standalone 5G NR, which implies full
user and control plane capability for 5G NR, are expected to be
completed in June 2018 as part of 3GPP Release 15.
Plans for designating The spectrum needs of 5G can be segmented into three key
and releasing frequency ranges, <1 GHz, 1-6 GHz, and >6 GHz, reflecting the
spectrum for 5G
desire to deliver widespread coverage and capacity. Out of sub-1
GHz spectrum, the 700 MHz band is frequently identified as a
suitable candidate band and would be demonstrative of the
transition from 4G to 5G. Within the 1-6 GHz range, the first band
prioritised for 5G is the 3.6 GHz band. For the third category, the 26
GHz has been identified as a pioneer band and generally millimetre
wave (mmWave) spectrum is being considered for meeting the
needs of transmitting large amounts of data. World Radio
Conference (WRC) 2019 will be vital to realise the ultra-high-speed
vision for 5G. The work at WRC-19 will look at spectrum for mobile
broadband in frequencies between 24.25 and 86 GHz.
Global initiatives and All the major economies are vying for leadership in 5G and to make
events to support 5G the first launch. South Korea aims to complete the deployment of a
deployment
commercial 5G mobile network in the second half of 2019. A 5G trail
service was launched by 2018 Winter Olympics10 on the KT mobile
9
We acknowledge that this band (which includes 3400-3800 MHz) is referred to in a
number of different ways. For example, EC refers to this as the 3.5 GHz band, RSPG
refers to it as the 3.6 GHz band and CEPT refers to the 3.4 – 3.8 GHz band. For the
consistency throughout this report we refer to the 3.6 GHz band.
10
International Olympic Committee, “Fans of the Olympic Winter Games 2018 to
experience word’s first broad-scale 5G network”, 9 February 2018. Available at:
https://www.olympic.org/news/fans-of-the-olympic-winter-games-2018-to-
experience-world-s-first-broad-scale-5g-network
9
network (with Intel and Samsung) to support the high data usage
arising from the combination of the ‘Olympic App’, drones, virtual
reality systems, large stadiums and promotion of social media
sharing throughout the games. China11 and Japan (2020 Tokyo
Olympics)12 are planning commercial deployments by 2020, with
the UK also running a 5G testbed and trials programme13 targeting
2020 for commercial deployments. In France, ARCEP opened a “5G
Pilot” window, to allow all of the players along the 5G value chain to
explore use cases and the challenges of this new generation. The
scheme seeks to bring together different players along the value
chain, allocate frequencies to interested players, conduct the 5G
trials and obtain initial feedback on the potential uses of 5G.14
The European The European Commission (EC) is applying considerable effort to
Commission has set place Europe at the forefront of 5G. To ensure the timely
targets and
deployment of 5G, the EC has requested Member States to align
milestones
roadmaps, make provisional spectrum available, undertake trials
and promote early deployment. The EC has set out a preliminary
timetable for achieving widespread 5G deployment in Europe:
• The first milestone is the launch of pre-commercial 5G trials
in early 2018. This should be followed by the launch of early
5G networks in the same year;
• The EU has set the ambitious target to launch fully
commercial 5G services by the end of 2020. To this end, the
EC has requested member states to make pioneer bands
available ahead of the WRC-19;
• The EC also asked member states to identify at least one
major city to be 5G enabled in 2020, while all urban areas
and major terrestrial paths should have uninterrupted 5G
coverage by 2025.15
Furthermore, to accelerate research and development of 5G, the EC
is co-financing an initiative called the 5G-infrastructure public
private partnership (5G-PPP), which was launched in 2013. 5G-PPP
brings together a wide range of industry stakeholders who, through
11
Mobile World Live article “China outlines 5G R&D roadmap”. Available at:
https://www.mobileworldlive.com/featured-content/top-three/china-outlines-5g-
rd-roadmap/
12
GSMA 5GMF – “Strategy and Activities on 5G Development in Japan and 5GMF”.
13
UK DCMS 5G Testbed & Trials Programme-
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/6
72118/Next_Generation_Mobile_Technologies__An_Update_to_the_5G_Strategy
_for_the_UK_Final_Version_with_Citation.pdf.
14
Arcep, “5G, Frequencies And Innovation”, 16 January 2018. Available at:
https://www.arcep.fr/index.php?id=8571&no_cache=1&no_cache=1&tx_gsactualit
e_pi1[uid]=2119&tx_gsactualite_pi1[backID]=26&cHash=b9046864c82ce08ebebc
240f271ef97f&L=1
15
European Commission Communication, “5G for Europe: An Action Plan” and the
accompanying “Staff Working Document”, 14 September 2016.
10
research projects and cross-project working groups, will help deliver

infrastructures, technologies and standards to allow 5G deployment
from 2020. Within the 5G-PPP, the 5G Infrastructure Association (5G-
IA) represents private interests, with the EC representing public
interests.
Projects to support 5G-PPP projects and working groups have been organised into
specification and three phases. Phase 1 has been completed and was focused on
trails for 5G
specification. Phase 2 projects will actively contribute to the
technologies and
use-cases development and testing of 5G technologies for vertical use cases.
Phase 3 will target large-scale trials through projects focusing on
end-to-end test facilities and platforms and on vertical pilots
integrating 5G technologies.
5G-PPP’s Phase 2 and Phase 3 projects are typically Pan-European
Trials. The 5G-PPP is working on a comprehensive strategy to
develop Pan-European coordinated trials, as well as international
trials with non-EU partner countries, addressing several key
elements of the EU 5G Action Plan (5GAP)16. The 5G-PPP ”5G Trials
Working Group” released a “5G Pan-European Trials Roadmap”
earlier in 2017, establishing four main pillars, including 5G Private
Trials, 5G Vertical Pilots, 5G for Euro 2020 and 5G Trials Cities17.
2.4 Regulatory environment

It is expected that regulation will play a key role in shaping the roll-
out of 5G. Not only must regulators make suitable spectrum
available in the appropriate bands, in sufficient amounts and with
appropriate licence conditions, but there may be other regulatory
decisions that may help or hinder the roll-out of 5G. As a general
principle, we can expect that the significant investments required
for 5G will be encouraged by a stable, consistent and proportionate
regulatory framework for all stakeholders.
Whilst we consider the role of regulation in more detail later in this
report, it is helpful to understand the key decisions already made at
a European level and the plans already in place to facilitate 5G
investment, as well as some of the topics relevant to 5G currently
being investigated by National Regulatory Authorities (NRAs).
16
A roadmap by European Commission that sets out measures to guarantee a
coordinated approach among all member states to make 5G accessible in line with
the 5GAP goals of “at least one major city [per member state] to be 5G enabled in
2020”, and making sure that “all urban areas and major terrestrial paths (…) have
uninterrupted 5G coverage by 2025”.
17
Defined in 5G IA Pan-European trials roadmap version 1.0. Available at:
https://5g-ppp.eu/presentation-of-the-5g-pan-european-trials-roadmap/
11
EU-level identification of spectrum for 5G

Europe has largely agreed on some important steps that will need
to be undertaken to support the roll out of 5G infrastructure and 5G
services, the first of which is to ensure the EU-wide availability of 5G
spectrum for coverage and capacity uses as soon as possible.
For example, priorities for 5G spectrum harmonisation and
identification of ‘pioneer bands’18 for 5G in Europe were outlined in
“5G for Europe: An Action Plan”19 (5G Action Plan). The action plan
and accompanying staff document set out a strategy aimed at
ensuring that all urban areas have uninterrupted 5G coverage by
2025, and the EC proposes a coordinated approach to 5G
deployment in Europe.
The 700 MHz, 3.6 To achieve this goal, the EC action plan identified the 700 MHz and
GHz and 26 GHz the 3.6 GHz20 bands for early 5G deployment. Following a request by
bands have been
the EC to publish an opinion on 5G spectrum requirements, in its
identified as key
bands for 5G “Strategic Roadmap Towards 5G for Europe” the Radio Spectrum
Policy Group (RSPG) agreed that the 3.6 GHz band is most suitable
for the immediate introduction of 5G and proposed to Member
States that large blocks of this band ought to be made ready for
authorisation by 2020.21 The 700 MHz band is recommended for
nationwide and indoor coverage22. The RSPG also recommends the
26 GHz (24.25-27.5 GHz) band as a pioneer band above 6 GHz,
suggesting that a sufficiently large portion (e.g. 1 GHz) of the band
ought to be made available by Member States in response to local
market demand by 2020.23 The opinions also outline
recommendations regarding licensing and flexibility of
authorisation for these bands.
These findings and recommendations are also in line with the EC
mandate issued to The European Conference of Postal and
Telecommunications Administrations (CEPT) requesting a report on
18
Pioneer bands are described as early available frequency bands to be
harmonised for 5G use.
19
European Commission Communication, “5G for Europe: An Action Plan” and the
accompanying “Staff Working Document”, 14 September 2016.
20
We acknowledge that this band (which includes 3400-3800 MHz) is referred to in
a number of different ways. For example EC refers to this as the 3.5 GHz band, RSPG
refers to it as the 3.6 GHz band and CEPT refers to the 2.4 – 3.8 GHz band. For the
consistency throughout this report we refer to the 3.6 GHz band.
21
Radio Spectrum Policy Group, “Strategic Roadmap Towards 5G for Europe –
Opinion on spectrum related aspects for next- generation wireless systems”, 9
November 2016 and Radio Spectrum Policy Group, “Strategic Roadmap Towards
5G for Europe – RSPG Second Opinion on 5G Networks”, 30 January 2018.
22
Radio Spectrum Policy Group, “Strategic Roadmap Towards 5G for Europe –
Opinion on spectrum related aspects for next- generation wireless systems”, 9
November 2016.
23
Radio Spectrum Policy Group, “Strategic Roadmap Towards 5G for Europe – RSPG
Second Opinion on 5G Networks”, 30 January 2018.
12
technical conditions, as well as sharing conditions for 5G

spectrum.24
The CEPT is mandated to develop harmonised technical conditions
for spectrum use in the 26 GHz band. The CEPT has also been
requested to revise harmonised technical conditions in the 3.6 GHz
band, and to study potential extensions of the 1.5 GHz band.
Both the RSPG opinion and the CEPT mandate have emphasised the
importance of taking into account international developments to
facilitate interoperability and economies of scale from common
tuning range. The mandate also indicates that the 700 MHz, 800
MHz, 900 MHz, 1.5 GHz, 1800 MHz, 2.6 GHz and 3.6 GHz bands are
already potentially available for 5G use.25
66-71 GHz might Although on the agenda for WRC-19, the 32 and 42 GHz bands are
meet future needs not mentioned in the EC mandate. Even though RSPG identified the
that exhaust the 26
32 GHz band as a priority for studies in its first opinion, it is now of
GHz band
the opinion that 31.8-33.4 GHz band should no longer be
considered a priority because of reasons such as incompatibility for
radio-navigation, declining interest, and interest in keeping this
band for backhauling fixed services. The RSPG is also of the opinion
that there is no urgency in potential harmonisation of the 42 GHz
band because the 26 GHz band will eventually provide capacity in
excess of 3 GHz of spectrum for 5G, which is likely to cover the
potential demand. Further, in its second opinion, RSPG suggests
that the 66-71 GHz band should be prioritised in terms of studies for
second-stage 5G mmWave bands for reasons such as the reported
lack of use of these bands in most CEPT countries, its proximity to
and better propagation characteristics than the adjacent 57-66 GHz
band, its existing primary allocation for mobile and its potential to
become a primary European band.
Therefore, at present, the three bands that have been identified for
harmonisation and to facilitate the launch of 5G in Europe are:
• The 700 MHz band for sub 1 GHz spectrum;
• The 3.6 GHz band for spectrum between 1 GHz and 6 GHz;
and,
• The 26 GHz band for above 6 GHz spectrum.
24
EC mandate to CEPT to develop harmonised technical conditions for spectrum
use in support of the introduction of next-generation (5G) terrestrial wireless
systems in the Union, 7 December 2016.
25
RSPG also recognises the need to ensure that technical and regulatory conditions
for all bands already harmonised for mobile networks are fit for 5G uses.
13
Support for investment in 5G

In 2016, in response to a call for input for the 5G Action Plan, a
number of telecommunication companies wrote a manifesto
formulating their recommendations for 5G deployment.26
Operators have The manifesto asserted that it would not be viable to invest in 5G
claimed that networks without changes to the regulatory environment and
regulation needs
argued that, in order to incentivise investment, MNOs would need
changing to support
5G investment greater regulatory certainty through:
• simplification of infrastructure access regulation;
• encouragement of co-investment and risk-sharing models;
and
• withdrawal of ex-ante regulation on 5G-ready infrastructure.
Furthermore, the manifesto claimed that 5G’s need for dense
infrastructure deployment could only be achieved through removal
of deployment barriers. For example, network slicing is a
fundamental functionality of 5G to bring added value across vertical
industries yet, according to the manifesto, the Net Neutrality
guidelines put forward by BEREC are excessively prescriptive and
might limit the telecom industry’s motivation to invest in 5G by
creating significant uncertainties around returns on 5G investment.
The draft ECC looks Some of the issues raised in the operators’ manifesto are covered in
at measures to the proposal for a European Communications Code (ECC)27. The
encourage 5G roll-
out
draft ECC sets out proposals that support 5G roll-out by creating a
more favourable investment environment.
In particular, small cell deployment is intended to be facilitated by
the proposals of Article 55 and 56. Article 55 intendeds to clarify
conditions for relating to the use of RLANs (e.g. Wi-Fi) within public
communications networks, including the use of equipment such as
routers on customers’ premises to support publicly accessible
networks and the ability of end-users to maintain control over their
own RLANs. Article 56 intends to allow deployment and operation
of small-area wireless access points, thus reducing deployment
costs for very dense networks.
Articles 45 and 46 proposed in the draft ECC also intend to enable
more flexible management of spectrum rights by competent
26
The Manifesto is effectively an open letter to the Commissioner of the Digital
Economy and Society dated 7 July 2016. Its signatories are the telecoms operators
BT, Deutsche Telekom, Hutchison Whampoa Europe, Orange, Proximus, KPN, Tele2,
Telecom Italia, Telefonica, Telekom Austria, Telenor, Telia, Vodafone; the vendors
Ericsson and Nokia and the satellite operators, Inmarsat and SES. Five companies
from vertical industries expressed interest in the initiative.
27
Proposal for a Directive of the European Parliament and of the Council
establishing the European Electronic Communications Code, COM/2016/0590 final
- 2016/0288 (COD), 14 September 2016
14
authorities, which will be more suitable for 5G deployment. These

articles promote general authorisations over individual licences and
determine that individual spectrum rights should be minimised and
only granted where necessary to maximise efficient use. The Articles
encourage shared use of spectrum and the trading and leasing of
frequency use rights. However, we understand that these proposals
are still under discussion and subject to change.
In their manifesto, the operators also stressed the need for seed-
corn funding to drive 5G deployment. One such instrument was a
fund to incentivise verticals to undertake large-scale trials with
innovative 5G services; it was suggested that such trials could help
align requirements between mobile operators and verticals. The 5G-
PPP confirmed this is a crucial step towards 5G implementation28
and has integrated this recommendation into its roadmap, as
described above.
The manifesto also recommends the EC to set up a 5G Venture Fund
that supports start-ups in developing 5G technologies and
applications. The telecoms companies warned that any
“infrastructure funding should be focused on physical infrastructure,
such as ducts or digital spine”. In the 5G Action Plan, the EC took into
consideration the suggestion made in the manifesto for a 5G
Venture fund and proposes to work with the industry and EIB group
to assess the possibilities.
Significant funding is In any case, with regards to funding development and trials, the EC
already committed has committed €700 million in funding to the projects selected by
to trials and
the 5G-PPP, which is expected to leverage private investment of five
demonstrations
times that amount.29 The Phase 3 projects, which will run from 2018
to 2020 in accordance with the EC roadmap, will be leading with
large-scale trials and demonstrations of 5G. In addition to these
trials using pioneer bands, the 5G-PPP has expressed a need for
trials of new frequency bands available from 2020 before their
network deployment.
Consultations at a national level

In addition to discussions and regulatory guidance of various forms
issued at the European level, a number of NRAs have individually
launched calls for input (CFI) or consultations to help shape their
thinking on how to support and encourage 5G deployments.
28
5G-PPP white paper, “Vision on Software Networks and 5G SN WG”, January 2017.
29
European Commission, “EU unveils 22 projects selected under the 5G Public,
Private Partnership”, 1 July 2015. http://europa.eu/rapid/press-release_MEMO-13-
1159_en.htm
15
Ofcom, UK For example, the UK regulator Ofcom has recently published its
action plan to enable 5G in the UK, which provides an overview of
the spectrum pipeline to meet the increasing demand for mobile
broadband.30 In order to enable a range of players to test innovative
new uses, Ofcom has also launched a new trial and innovation
portal.31
Ofcom has already indicated the 700 MHz band will be auctioned in
201932. An auction including frequencies for 5G deployment within
the range 3.4-3.6 GHz is scheduled to take place in 201833. Ofcom
have also announced that further spectrum in the range 3.6–3.8
GHz will be made available for mobile and auctioned in 2019.34
Following an initial Call For Input on 3.8-4.2 in 2016, Ofcom plans to
further consider the potential for increased shared access to this
band for innovative new uses, while taking into account existing
users.
In 2017, Ofcom also launched a call for inputs (CFI) on likely
demand, timelines and spectrum authorisation options for 5G at 26
GHz35. Ofcom’s CFI was focussed on:
• The likely demand, with regard to locations, services, channel
bandwidth and deployment models to use 5G technologies at
26 GHz;
• The timelines for 5G equipment operating across the 26 GHz
band and the technology features that may be relevant to
authorising spectrum;
• The range of spectrum authorisation options that may be
relevant in response to the specific market demand at 26 GHz
and whether different authorisation types are required to meet
the needs of different deployment models and services;
• Options for existing users of the band to allow the introduction
of 5G.
Ofcom also sought stakeholders’ views on making the 26 GHz
pioneer band available on a progressive basis. The UK regulator also
30
Ofcom “Enabling 5G in the UK”, 9 March 2018. Available at:
https://www.ofcom.org.uk/__data/assets/pdf_file/0022/111883/enabling-5g-
uk.pdf
31
Ofcom, “Innovation licensing including 5G”. Available at:
www.ofcom.org.uk/innovation
32
Ofcom, “Update on spectrum in the UK”, 8 February 2017.
33
Ofcom, “Award of 2.3 and 3.4 GHz spectrum by auction”,
https://www.ofcom.org.uk/spectrum/spectrum-management/spectrum-
awards/awards-in-progress/2-3-and-3-4-ghz-auction
34
Ofcom, “Improving consumer access to mobile services at 3.6GHz to 3.8GHz:
Update on timing of spectrum availability”, 2 February 2018, Available at:
https://www.ofcom.org.uk/__data/assets/pdf_file/0018/110718/3.6GHz-3.8GHz-
update-timing-spectrum-availability.pdf
35
Ofcom, “Call for inputs on 5G spectrum access at 26 GHz and update on bands
above 30 GHz”, 28 July 2017.
16
recommends studying the possibility of using the 40.5-43.5 GHz (as

part of a wider tuning range at 37-43.5 GHz) and 66-71 GHz bands
for 5G.
RTR, Austria RTR, the Austrian regulator, has confirmed it will make the 700 MHz
band available for 5G in 2020 at the latest, in line with the EC’s
request.36 In their spectrum release plan, the regulator indicated
that the 700 MHz band is likely to be auctioned together with the
1.5 GHz band.
RTR has already started preparations for a 5G spectrum auction
planned to run in 2018. The spectrum to be awarded consists of
frequencies in the 3.6 GHz band. Licensees will be allowed to each
use one frequency block for 5G deployment. Spectrum in the 3.4-3.6
GHz range will be available from end of 2019, while the 3.6-3.8 GHz
range will be available in 2020. RTR plans to allow infrastructure
sharing to facilitate densification of small cell deployment for 5G
roll-out. The regulator also considered spectrum sharing through a
two-tier model that allows secondary use of frequencies as long as
this does not interfere with the licence holder’s use.37 However, we
understand that this would require an amendment of the telecoms
act that has not yet been made.
BNetzA, Germany In 2017, the German regulator BNetzA launched a public
consultation on 5G frequencies to identify demand for 5G
spectrum.38 The 3.6 GHz band and also the 2 GHz band were subject
to consultation over likely demand (despite the 2 GHz band not
having been identified as one of the pioneer bands).
Ministerie van The Netherlands’ Ministry of Economic Affairs and Climate Policy is
Economische Zaken preparing for a national frequency auction of 4G/5G mobile
en Klimaat,
spectrum in the 700 MHz, 1400 MHz and 2100 MHz bands in
Netherlands
autumn 2019. The auction will be organised on the basis of the
‘Nota Mobile Communication’39 white paper to be published later
this year, after the Ministry has finished processing the consultation
responses to an earlier draft version.
Additionally, in 2018 the Netherlands government will evaluate the
potential for opening up the 3.6 GHz band to mobile services. The
3.6 GHz band is currently restricted to military usage in certain areas
of the country.
ComReg, Ireland The Irish regulator (ComReg) has proposed to award blocks in the
26 GHz band for fixed links, as current fixed link licences are due to
36
RTR, “Spectrum Release Plan”, December 2016
37
RTR, “Consultation on the 3.4-3.8 GHz award procedure”, 28 July 2017.
38
Bundesnetzagentur, “Key Elements for the rollout of digital infrastructures and
Identification of Demand for nationwide assignments in the 2 GHz and 3.6 GHz
bands”, 2017.
39
Ministerie van Economische Zaken en Klimaat, “Nota Mobiele Communicatie
2017”, 2017.
17
expire shortly. However, 10-year licences are proposed – shorter

than the typical duration of licences – specifically so that future
alternative uses, such as 5G, are not precluded given likely realistic
timescales for 5G deployment.40 ComReg notes that only about one-
quarter of the entire 26 GHz band would be awarded in this process,
leaving in excess of 1.2 GHz of currently unused spectrum available
at the top of the band to meet future 5G needs.41
The 3.6 GHz band has already been awarded in Ireland.42 Winners
included the incumbent mobile network operators, an FWA
operator - Imagine, who obtained spectrum only in rural areas - and
a new entrant, Airspan. On the basis of its website, it appears that
Airspan is interested in supplying wholesale network services to
other operators.
Arcep, France Arcep launched a public consultation on 5G in January 2017 on
future uses of several frequency bands, including the 2.6 GHz and
3.6 GHz bands, and on the methods used to allocate these bands.
Arcep will take an inventory of stakeholders’ needs, identify the
frequency bands best suited to each requirement, and explore
appropriate spectrum allocation methods. The responses to
consultation will be considered alongside the lessons learned from
a number of trials underway in the 2.6 GHz and 3.6 GHz bands to
determine how the 2.6 GHz and 3.6 GHz bands could contribute, in
the short term, to ultrafast fixed access and PMR network rollouts
and, in the medium term, to deploying 5G.43
MINETAD, Spain In July 2017, Spain’s Ministry of Energy, Tourism and Digital Agenda
(MINETAD) launched a public consultation regarding the future
introduction of 5G technology. Input is sought with regard to the
current status of 5G technology, potential 5G services and
applications, network deployment, radio spectrum, 5G prototypes
and 5G research and development (R&D). Following the
consultation, in December 2017, MINETAD released the 5G Plan for
the 2018-2020 period.44 The National Plan 5G aims to promote early
implementation of networks 5G in Spain, as well as to contribute to
the development of R&D in this area.
Within this Plan, Spain’s Ministry for Energy, Tourism and the Digital
Agenda has announced a series of measures designed to promote
40
ComReg, “Response to Consultation and Draft Decision on the Proposed 26 GHz
Spectrum Award”, 23 February 2018, §4.82.
41
Ibid, §2.26.
42
ComReg, “Five Winning Bidders in ComReg’s 3.6 GHz Band Spectrum Award”, 22
May 2017. Available at: https://www.comreg.ie/five-winning-bidders-comregs-3-6-
ghz-band-spectrum-award/
43
Arcep, “Arcep launches a public consultation on “New frequencies for the
regions, businesses, 5G and innovation””, 6 January 2017.
44
MINETAD, “Plan Nacional 5G 2018-2020”, 1 December 2017.
18
the deployment of 5G throughout the country, including a tender in

the 3.6 GHz and 1.5 GHz bands “in early 2018”.
19
Overview of key 5G technologies
3 Overview of key 5G technologies

Although the final requirements for 5G have not been finalised
yet45, there is common agreement on the main families of usage
scenarios and applications that 5G might support.46 The three main
uses are illustrated below:
• Massive mobile connectivity, that would enable enhanced
mobile broadband (eMBB);
• Connectivity of millions of devices, that would enable massive
machine type communication (mMTC); and
• Resilient, instantaneous connectivity, that would enable ultra-
reliable and low latency communications (URLLC).
Figure 3: The main families of usage scenarios and applications 5G might
support
MASSIVE MACHINE-TYPE
COMMUNICATIONS (mMTC)
u  Adoption and utilization across

ENHANCED MOBILE different sectors on a massive
BROADBAND (eMBB) scale
u  Uptake of mobile technologies

u  Extension of the existing to address MIoT applications
4G value proposition
u  Extending cellular
coverage
u  Improving network ULTRA-REALIABLE, LOW-

capacity LATENCY COMMUNICATIONS
(UR-LLC)
u  Applications requiring high

reliability and low latency
connectivity, e.g. mission
critical services
u  Strong security and availability

requirements
To effectively satisfy these use cases, 5G must combine a number of

new technical capabilities and technologies. Therefore, 5G is far
more than a faster radio access technology; it involves fundamental
changes in network architectures.
45
However, in December 2017, 3GPP announced the completion of Release 15 NSA
5G NR specification, which introduced a relevant milestone to complete the first
implementable specification for Non-Standalone 5G NR operation.
46
identified by ITU-R M.2083 ¨IMT Vision – Framework and overall objectives of the
future development of IMT for 2020 and beyond¨.
20
3.1 Technical capabilities that differentiate 5G

ITU-R has defined eight key technical capabilities for IMT-2020,
which are summarised in the figure below:
Figure 4: Technical capabilities of IMT-2020
1- 20 Gbps 1- 10 ms
Peak data
Total amount of traffic handled L atency Round trip time for a packet of
rate
by a single cell data
User 10 - 100 Mbit/s 10k - 1million devices/km 2

C onnection
ex perienced Total amount of traffic Number of devices fulfilling a
density
data rate experienced by the end- user certain QoS
90% more efficient

15 - 30 bit/s/H z N etwork
Peak spectral C apability of a RIT (radio
Information rate that can be energ y
efficiency interface technology) to
transmitted efficiency
minimize energy consumption
350- 500 km/h 0.1– 10 Mbit/s/m 2

Area traffic
Mobility Maximum mobile station speed Total traffic throughput served
capacity
at which certain QoS is achieved per geographic area
Source: Report ITU-R M.2410-0 (11/2017) Minimum requirements related to technical performance for IMT-2020
radio interface(s).
IMT-2020 defines an Technical capabilities under IMT-2020 present a significant

envelope of technical enhancement vis-à-vis its predecessor (IMT-Advanced – sometimes
possibilities within
which trade-offs can
referred to as 4.5G). The figure below shows the improvements in
be made terms of peak data rates, spectral efficiency, mobility, latency,
energy efficiency and connection density, amongst others. For the
avoidance of doubt, there are trade-offs and not all network
performance parameters can be pushed to the limits of the IMT-
2020 capabilities simultaneously. However, a key aspect of 5G is
that virtual networks can be created prioritising certain features
(e.g. low latency or energy efficiency).
21
Figure 5: Enhancement of key capabilities from IMT-Advanced to IMT-2020

Peak data rate
(Gbit/s) ed
nc
rie te
Ar cap it/s
e
ea ac /m
p a
ex a r /s)
(M
tr ity 2 )
er at bit
affi
b
Us d (M
c
20
0
10
10
1
10
0.
1
100x 1x 3x Peak spectral
Network energy
efficiency
efficiency 1x (bit/s/Hz)
35
0
5
10
6
10 10
50
0
1
M Km
ic sit ion
ob /h
(
ev en ect
ili )
2
es y
ty
m
d n
/k
n
Co
Latency
(d
(ms)
IMT-Advanced
IMT-2020
Source: Recommendation ITU-R M.2083-0
Although the final requirements for 5G have not been finalised yet,
each of the three main families of usage scenarios (i.e. eMBB, URLLC,
mMTC) are expected to have differing needs in terms of these
technical capabilities, as shown in the figure below.
Figure 6: Enhancement of key capabilities from IMT-Advanced to IMT-2020
Peak data rate
(Gbit/s)
Peak spectral
N etwork energy
efficiency
efficiency
(bit/s/H z )
Latency
(m s)
Enhanced mobile broadband (emBB)
Ultra- reliable and low latency communications (URL L C )
Massive machine type communication (mMTC )
Source: Recommendation ITU-R M.2083-0
22
The charts above remind us that it is not possible to achieve peak

values across all capabilities simultaneously and a trade-off will have
to be made when defining a differentiated service to support a
particular use case. These differentiated services can be
implemented using network slicing to create a number of virtual
networks with different performance characteristics hosted on a
common infrastructure; within the overall constraints of that
infrastructure, service characteristics of particular network slices can
be varied agilely through software reconfiguration.
Defining 5G Under this scenario, the concept of 5G coverage is not as
coverage if straightforward to define as for previous generations of mobile
capabilities vary
technology. 5G services at a given location will not have fixed
from location to
location characteristics, but rather represent a choice made within the
overall range of possibilities defined by IMT-2020. In particular, we
expect to see geographic areas that meet only part of IMT-2020
targeted technical capabilities despite a formal 5G standard (e.g.
Release 15) being used to define what we mean by ‘5G’. Moreover,
the choice of capabilities available at a location might be altered
from time to time to meet changing user requirements.
For instance, there may be less densely populated areas that will be
covered using 5G standard aimed at providing mMTC capabilities
(e.g. to serve the needs of precision agriculture), but at the expense
of reaching the data rates available from eMBB. Another example
could be those areas that require significant upgrades to the
network (e.g. edge computing) to minimise latency, which may not
be required ubiquitously.
Massive system To support the dramatic surge in traffic loads for mobile
capacity communications systems in an inexpensive manner, 5G networks
are expected to supply data with much lower cost per bit
compared with the current networks. By deploying the most cost-
effective combination of access technologies, operators can build
massive capacity as and when it is needed. Operators will become
able to intelligently steer, spread and load balance traffic to the user
across multiple access technologies and theoretically achieve
capacity greater than 10 Gbps.
Very low latency Latency will be driven by the need to support new applications
involving traffic safety, control of critical infrastructure and industry
processes that would require very lower latencies. Future 5G
infrastructure is expected to reduce end-to-end latency by a factor
of 5, reaching 5ms (for instance, for tactile internet) and with radio
link latency to a target of 1ms or less (for instance, for real time
mobile control facilitated by mobile edge computing).
Very high data rates 5G should boost data rates by about 100 times with peak data
rates in the order of 10 Gbps. However, peak data rates (which are
the maximum attainable data rates under ideal conditions) are less
relevant than user-experienced data rates actually achievable under
real-life conditions. Ambitions for user-experienced data rates are:
23
• >10 Gbps data rates in specific scenarios such as indoor and

dense outdoor environments;
• >100 Mbps data rates should generally be realisable in urban
and suburban environments; and
• >10 Mbps data rates should be accessible nearly everywhere,
including rural areas.
Ultra-high reliability Reliability refers to the capability of transmitting a given amount of
and availability traffic inside a fixed time duration with a high success probability.
For critical services that 5G might cater to, loss of connectivity and
deviations in quality of service should be exceptionally rare. A few
industrial applications might even need a successful packet delivery
guarantee of within 1ms at a probability higher than 99.9999%. 5G
has the capability to allow terminal devices to be connected to
multiple cells simultaneously; to the extent that failure risks at
closely located cells are not strongly correlated, this should provide
a reliability improvement.
Very low device cost To enable the vision of billions of wirelessly connected sensors and
and energy devices, it is important to focus on lowering the device cost and
consumption
energy consumption. 5G will enable low complexity receivers,
even when scaling to higher bandwidths, lowering the device costs.
It should be possible for 5G devices to be available at very low cost
and with a battery life of over 10 years.
Energy-efficient While energy consumption is a priority on the device side, energy
networks efficiency is emerging as an important requirement on the network
side as well. Network energy efficiency is the capability of a radio
interface technology to minimise the radio access network energy
consumption in relation to the capacity of traffic provided. Energy
efficient networks will lower operational costs. They may even
enable off-grid network deployments that are powered by solar
panels, helping to reach the most remote areas.
3.2 Enabling 5G access using a combination of

technologies
The front-running technologies and techniques that could
potentially enable 5G networks are presented in Table 7 and
explained in turn below.
24
Figure 7: Key technologies to enable 5G

Challeng es inherent to 4G
• Large number of base • Unable to transmit

• Less bandwidth • Low spectral efficiency
stations and receive at the
• N eed for new • Less uniform user experience
• N eed for more speed same time over same
spectrum • Low coverage capacity and throughput
• S low deployment frequency
Millim eter Waves Massive MIMO Full D uplex
new unused spectrum between high efficiency, throughput, transmit and receive data at
30-300 Ghz, increase capacity inexpensive, reduced latency, same time and frequency,
and speed fine spatial focusing doubles network capacity
high absorption, low reach causes interference, large interference, reciprocity issue
number of antennas required
Sm all Cells Beam form ing

cost effective capacity and coherent and focused data
coverage, travel through stream with reduced
obstacles, less power usage interference
high interference, backhaul, expensive, complicated, high
large number of cells required power requirement
N ew challeng es
Additional technolog ies to solve the rem aining challeng es
Currently wireless networks are typically spectrum constrained in

geographical areas where there is a high density of users. This has
resulted in less bandwidth for users, leading to a slower service and
more dropped connections. A solution possible with 5G would be
to use new spectrum that has never been used for mobile services –
mmWave. This will massively increase the capacity of the networks,
but has limited propagation47 and is unable to travel through
buildings and obstacles. To overcome this, carriers could install
thousands of miniature base stations - small cells. Such a radically
different network structure would provide more targeted and
efficient use of spectrum.
Furthermore, two technologies that would prove critical in enabling
5G are massive MIMO and beamforming. By featuring dozens of
antennas in a single array, massive multiple-input multiple-output
(massive MIMO) could increase the capacity of mobile networks by
a factor of 22 or greater48. Beamforming would allow a large
47
Millimetre waves are known to have much lower coverage capacities due to
higher absorption, which will be a challenge for telecom operators. Industry is
currently struggling to model, understand and learn how to plan networks in cases
such as rain, snow, changing foliage due to seasons, etc. for which the effect is
negligible for bands currently used for mobile services.
48
Nokia, “Beamforming for 4.9G/5G Networks – Exploiting Massive MIMO and
Active Antenna Technologies.”
25
number of users and antennas on a massive MIMO array to

exchange more information at once by improving signal to noise
ratios. Moreover, it results in a coherent and focused data stream
that can reach larger distances thereby increasing capacity of the
cell towers in terms of number of subscribers served. While massive
MIMO provides increased spectral efficiency, a new technology, so-
called full duplex, is currently being researched. Although few field
trials have been conducted so far, this technology is expected to
increase throughput and reduce latency, both of which are critical
requirements for 5G. It could double the capacity of wireless
networks by enabling the transceiver to transmit and receive data at
the same time over the same frequency.
3.3 A single network to serve multiple demands

5G also has potential to bring benefits over previous radio network
technologies by allowing networks to be extremely agile and meet
a potentially wide range of different demands for very low latency,
bandwidth requirements, and device density (for example).
Stakeholders have identified key enablers that will help new
services with diverse requirements to be accommodated efficiently
and rapidly using a single network. These enablers are summarised
in the figure below:
Figure 8: 5G virtual end-to-end networks tailored to serve application requirements
Use cases
Manag em ent, Orchestration and Advanced Analytics

eMBB
Virtualiz ation Software D efined D istributed N etwork Slicing

N etworking Cloud
VN F VN F Applications
(SD N )
in Cloud
URLLC
m MTC
Virtualisation of Virtualisation renders network functions that typically run on

networks dedicated and specialised hardware as software that can be run as
virtual machines on commodity servers. Thus, it enables hardware
resources to be managed as a common resource pool. Network
function virtualisation (NFV) enables the creation of new network
functions on-demand.
26
Software Defined To make the most of virtualisation, software defined networking

Networking (SDN) (SDN) is able to adjust the network in software, enabling network
programmability and sequencing of functions. The convergence of
NFV and SDN will permit operators to be more responsive to
customer demands, both by being able to tailor bespoke services
(within reasonable time and cost) and through on-demand
provisioning by customers themselves via self-service portals.
Distributed Cloud Once networks are virtualised and made programmable, it becomes
possible to shift computing resources to the most appropriate
location. Some use cases might require low latency responses, for
which mobile edge computing (i.e. cloud computing resources
within the network itself located close to the user) may be
appropriate.
Network Slicing Network slicing will enable operators to deliver networks on an as-
a-service basis and meet the wide range of use cases that the 2020
timeframe is expected to demand. With this technology, a single
physical network can be separated into multiple virtual networks
letting the operators to offer optimal support for the different
services for different user segments.
Management, This layer is expected to act as the glue that holds together the
Orchestration and
entire network. Once the network is enabled, it is important to
Advanced Analytics
deliver the service in an efficient way to the end user. This needs
smart management, smart orchestration and advanced analytics to
expose the network in an abstract way.
27
Use cases for 5G
4 Use cases for 5G

Much has been said in the public domain about the potential use
cases that the enhanced technical capabilities of 5G could support,
from improvements in mobile broadband services to futuristic
sounding services such as fully autonomous vehicles. However, the
realisation of 5G use cases will be a step-by-step process and many
potential uses are currently at very early stages of development;
many use cases that may eventually prove important once 5G is
widespread may not even have been identified yet.
Whilst we might expect that early 5G deployments will be focussed
on improving existing mobile broadband propositions, there may
be other use cases for 5G enabled by its distinctive new capabilities
over previous radio access technologies.
To this end, much of the public discussion of 5G to date has
considered eye-catching use cases in specific industries where 5G is
expected to trigger new business opportunities or bring significant
additional value to existing services. For example, 5G use cases are
often discussed in sectors such as automotive, healthcare,
manufacturing and agriculture (amongst others). The figure below
illustrates the potential verticals and horizontals (the equipment
manufacturers and connectivity providers) that attract the most
attention (together with some illustrative case studies under each).
As we discuss later in the report, connectivity providers may not
always be telecoms operators, as verticals may use private networks
or source a solution from equipment manufacturers. Therefore,
there may not always be a role for traditional telecoms operators
(such as MNOs) within the value chain.
Figure 9: Overview of proposed verticals and illustrative case studies
Equipment/network manufacturers
Connectivity providers
Media and Energy and

Automotive Manufacturing Logistics Agriculture Healthcare
entertainment utilities
u  Connected u  Live events u  Logistics u  Drone delivery u  Agriculture u  Smart u  Remote

cars u  Immersive tracking u  Connected drones metering monitoring
u  Infotainment media u  Robot control ports u  Precision u  Smart grids u  Telesurgery
services u  Augmented farming
u  Platooning reality (AR)
28
Use cases for 5G
Beyond eMBB, in the Whilst there be some particular use cases for which 5G availability
short-to-medium will be critical, we are somewhat sceptical that, at least in the early
term there may be
years, there will be use cases dependent on 5G that will create
few use cases entirely
dependent on 5G sufficient incremental revenue that this could significantly affect 5G
deployment incentives. Indeed, many use cases are unlikely to
emerge with any clarity in the short-to-medium term, due to lack of
demand, commercial risk, lack of suitable hardware, or simply due
to the large amount of research and development that must be
conducted beforehand.
Whilst the EU, and other countries, have in place a number of
initiatives to support research and promote trials, to date these
have focussed on showing what is technically feasible and
determining exactly how particular use cases may be supported by
technology and standards setting, but there has been limited
attention given to how (or whether) such services will be
commercially viable.
Commercial services When considering products and services that might be enabled by
and products 5G (and also improvements to companies’ internal processes that
wishing to use
might be enabled by 5G), there is clearly a high level of risk for
mobile connectivity
to provide a developers. It cannot be assumed that 5G will be ubiquitously
ubiquitous service available. It is reasonable to expect that - for a considerable time -
will likely use a mixed there will be a mixed connectivity environment, with 4G available
connectivity and Wi-Fi often providing indoor connectivity. Initial 5G
approach
deployment is likely to be focussed on urban areas. Therefore,
commercial services and products using mobile connectivity
wishing to provide a ubiquitous service will need to have fall-back
options rather relying entirely on 5G. We should expect that a
developer wanting to offer an application or service, or to embed
connectivity in a product, and market this across Europe, or even
globally, will want to reduce its exposure to 5G rolling out at
different rates in different countries by being able use whatever
networks are available.
5G will need to be The relevant alternatives will depend on the details of the
attractive relative to connectivity requirements (as we discuss below in a number of case
other technological
studies). Whilst 5G will provide significant enhancements and also
solutions
much more flexibility, where there are alternatives this will limit
what users will be prepared to pay additionally for 5G functionality.
Pricing models for 5G There will also need to be models that allow verticals to pay for 5G
will be an important connectivity in a manner that is efficient and does not create poor
factor in providing
incentives for the user. To take a specific example, a sensor network
the correct incentives
for adoption might benefit from using 5G due to the low power requirements
and the flexibility that using a public mobile network might provide
in locating those sensors and deploying them without needing any
additional infrastructure. However, payments to the network
operator need to reflect an appropriate balance between the
benefits to the user and the costs to the network operator. Paying
on a per device basis might be unattractive to the user and provide
incentives to reduce the density of its sensor network or switch to
an alternative way of connecting those sensors. However, if little
29
Use cases for 5G
data is transferred by each sensor, the marginal cost to the network

operator of additional sensors may be close to zero; nevertheless,
the network operator needs to price to recover the fixed costs of its
network infrastructure whilst trying not to discourage such users.
Therefore, to recover fixed costs, the network operator needs to
create a pricing structure that somehow reflects the benefits to the
user, which suggests that a pricing structure that charges a fixed
sum regardless of how many sensors were deployed at that location
would not be appropriate. However, there is a conflict, as charging
more to users with more sensors might cause some of those users
to switch to alternative technologies if they were cheaper at some
density.
This is just a hypothetical example, but it shows the general
problem of creating pricing menus for differentiated services that
both reflect the benefit that different users enjoy and yet does not
create poor incentives for users. These two aims cannot always be
easily aligned. As a result, there may be opportunities for individual
users to form partnership arrangements of various types with
connectivity providers – which might be a joint venture or even a
payment-by-results type of contract – in order to better align
incentives. In effect, the connectivity provider is integrating to some
degree into the services and products that embed that connectivity.
Our assessment of In this section, we consider some of the key sectors where the 5G
use-cases in the key debate has been focussed to date and consider the opportunities
verticals takes into
5G could bring. At the same time, we also consider the extent to
account the extent to
which alternative which some of these use cases could be supported by existing
technologies could technologies and how these alternatives could limit the incremental
be used to meet the value that 5G could bring in each of those cases.
requirements and
the enhancements We consider that it is very difficult to say at this stage whether any
5G might bring one sector/vertical as a whole, or any one specific use case, will or
will not need 5G. Many examples of use cases requiring wireless
connectivity are able to be served by alternative technologies.
Therefore, 5G will be competing with these alternatives. End users
will make adoption decisions based on cost effectiveness, additional
functionality and the flexibility that 5G provides. Many use cases
may simply evolve towards reliance on 5G as and when it becomes
available, using it as a complement to existing technologies.
For each sector, we seek to identify the specific requirements of that
sector and any potential issues that are likely to influence 5G
adoption. We will draw together the common features
subsequently.
4.1 Automotive
The automotive industry has been an early adopter of various
connectivity technologies. There has been a significant push by car
manufacturers in developing connected-ready cars, with a view to
30
Use cases for 5G
improving the in-car experience, helping traffic flow and overall

safety on the road, and obtaining information about the
performance of their vehicles and to assist maintenance. These are
small steps forward within a long-term vision of autonomous
vehicle control.
Opportunities for 5G
5G could bring improvements to services in the automotive
industry in the form of improved in-car ‘infotainment’ and
improvements to vehicle-to-everything (V2X) communications on
an evolution path to autonomous driving in the future.
Furthermore, as cars become more ‘connected’ and collect a larger
amount of data on location, usage, performance and other
telematics, opportunities open up for remote diagnostics, and
opportunities for developing emerging insurance models such as
‘pay-as-you-drive’.
Potential use cases
The wide range of use cases considered within the automotive
sector range from infotainment services such as on-demand
entertainment and trip advisory services in the vehicle, traffic
management, weather and location-based services while the
vehicle is moving, to road-side assistance and use-based insurance
and services including enhanced GPS and collision avoidance (with
a view to autonomous vehicles in the future). It is possible that 5G
services will be used to complement vehicle-to-vehicle (V2V)
communications, for example, to add to the information available
to the car including interaction with mapping data and
infrastructure to provide information about the route and the
environment beyond the ‘line of sight’ of V2V communications.
Table 2 below provides an overview of these use cases.
31
Use cases for 5G
Table 2: Potential use cases in the automotive industry

Use cases Impact on industry Critical requirements Estimated value
Improvements in Vehicle- Improved safety Latency < 5 ms

to-everything
communications Connection density > 10,000 /km2
(supporting the move to Mobility > 200 km/h
autonomous vehicles) and
allowing for features such
as collision avoidance,
emergency braking,
intelligent traffic systems
Infotainment services New opportunities for the Data throughput > 10 Mbps
provision of in-car
entertainment and Connection density > 10,000 /km2
information services of Mobility > 200 km/h
value to the driver and
passengers
Platooning Improved safety and Latency < 5 ms

efficiency
Reliability > 99.99999%
Data collection Generation of large Data throughput > 10 Mbps

amounts of telemetry
data opens up Connection density > 10,000 /km2
opportunities for new Mobility > 200 km/h
charging and insurance
models
Remote monitoring and Efficiency improvements - -

predictive maintenance
Driver assistance - ‘see Improved safety Latency < 5 ms

through’ the front vehicle,
in-dash junction cameras Data throughput > 10 Mbps
(for HGVs for example) Mobility > 200 km/h
Source: DotEcon and Axon based on publicly available information
Enhanced capabilities of 5G that could support these use-cases

It may already be possible to meet some of these use cases with
existing technologies and others may not be 5G critical. For
example, predictive maintenance might not require real-time, low
latency exchange of data, but rather uploading/sharing of
information intermittently. However, other aspects – such as
features associated with fully autonomous driving that rely on more
than just vehicle-to-vehicle communications or the increasing
demand for in-car infotainment - can expect to see improvements
from 5G. Although some of these use cases have been
demonstrated to work using 4G (for example, some infotainment
services), 5G could support improvements, especially if demand
increases and requirements on device density, mobility, and data
throughput become increasingly important.
Although use cases seen as stepping-stones to fully automated
driving are often thought to require 5G, we must be clear that some
32
Use cases for 5G
of those services only require basic vehicle-to-vehicle

communication and are not necessarily reliant on advanced
wireless technologies. For example, dedicated Short-Range
Communication (DSRC - a short-to-medium range, two-way wireless
technology for V2V communication) and Intelligent Transport
Systems (ITS) based on a set of standards relying on IEEE 802.11p49
within the dedicated 5.9 GHz band can be used. These technologies
allow for direct communication between source and destination
endpoints and can operate independent of a cellular network and
do so in a fully distributed fashion.5051
However, these technologies focus on short-to-medium range V2V
communications which are subject to limitations of distance and in
communicating with cars or infrastructure that are further ahead52,
as might be needed, for example, for the sharing of expected paths
of all surrounding vehicles in an autonomous world. Therefore,
taking this approach may be insufficient if the requirements of the
‘connected car’ extend beyond simple vehicle-to-vehicle
communication and require communications between other factors
such as vehicle-to-infrastructure (V2I) and vehicle-to-pedestrian
(V2P) communication, typically referred to together at Vehicle-to-
everything communications (V2X). For example, vehicles may also
need to communicate with infrastructure to allow vehicles to
receive and send information to traffic lights, roadside signs or
traffic control centres; and with other road users such as pedestrians
or cyclists etc. Depending on the type of data being transmitted and
the need for computations, information may also need to be
exchanged with a backend server for processing (such as
aggregation with other information and summarisation).
For V2X communications, one increasingly relevant solution is
Cellular-Vehicle-to-Everything (C-V2X), which provides a solution for
integrated V2V, V2I and V2P in operation with V2N by leveraging
existing cellular network infrastructure. C-V2X allows for two modes
of communication: a direct vehicle-to-vehicle mode and a network
communications interface for vehicle-to-network communication
via mobile networks.
49
IEEE 802.11p represents an amended version of the IEEE Std 802.11 for wireless
local area networks (WLANs) and specifies the extensions for the provision of
wireless communications while in a vehicular environment. See:
https://standards.ieee.org/findstds/standard/802.11p-2010.html.
50
5G-PPP White Paper on Automotive Vehicle Sectors.
51
We understand that from 2019 the Volkswagen group will use the WLANp
standard for communications between vehicles. See:
https://www.volkswagenag.com/en/news/2018/02/volkswagen_group_rapid_roa
d_safety.html
52
Arthur F. Little for Vodafone Group Plc, “Creating a Gigabit Society – The role of
5G”. Available at: https://www.vodafone.com/content/dam/vodafone-
images/public-policy/reports/pdf/gigabit-society-5g-14032017.pdf.
33
Use cases for 5G
The global cellular standard is now LTE-4G, which is seen as an

essential foundation to powering C-V2X. However, 5G offers a
further (and improved) option for C-V2X communication and is seen
to be more ‘future proof’ in the shift towards fully autonomous
driving in future. For example, although the 5G Automobile
Association (5GAA) acknowledges that forms of C-V2X technology
are available at present, they indicate that “…to support the
autonomous vehicles of tomorrow, the technology must evolve to meet
more demanding safety requirements. 5G will facilitate this evolution.
Its extreme throughput, low latency, and enhanced reliability will allow
vehicles to share rich, real-time data, supporting fully autonomous
driving experiences.” 53
Expected role of the automotive industry in driving 5G
development
According to the latest Ericsson ‘5G Readiness Survey’54 the
automotive sector is one of the most important industry sectors
that telecoms operators are focusing their 5G activities on, second
only to the Media and Entertainment sector.
By design, the C-V2X standard includes the possibility of direct
communication rather than communicating via networks due to the
high safety requirements and the high risks of being over-reliant on
a network that may have poor coverage in some areas; therefore,
vehicles would not be solely reliant on 5G mobile networks. There is
also demand for cellular connectivity for the provision of improved
infotainment services and performance diagnostics in the near
future,55 irrespectively of the rate of development of future 5G
networks. Therefore, we expect that this industry will be an early
adopter of 5G, but:
• for infotainment services 5G will represent an evolution of
services already provided over existing 4G cellular networks;
• for C-V2X communications, 5G will complement existing
cellular and short-range communications between vehicle
and infrastructure.
• 5G might also support some features of fully autonomous
vehicles in the long-term. However, the technology is
unlikely to be solely reliant on 5G given the safety aspects
and the need for 100% coverage and service guarantees.
In this context, the extent to which 5G will be adopted in the
automotive sectors will depend on sufficient coverage and an
53
5GAA, “Paving the way towards 5G”. Available at: http://5gaa.org/5g-
technology/paving-the-way/
54
Ericsson, “5G Readiness Survey 2017 – An assessment of operators’ progress on
the road to 5G”
55
In fact, a large number of vehicle manufacturers are already providing cellular
modems to, for example, allow the cars to act as Wi-Fi-hotspots.
34
Use cases for 5G
alignment of incentives between the equipment manufacturers

who will ‘embed’ the connectivity technology within the vehicles
and the provider of the mobile network. This raises a number of
interesting questions:
• Who will support the network investment that would be
necessary for automotive sector to be able to rely on
widespread 5G coverage?
• Will there be bilateral deals between individual
manufactures and telecoms operators to provide a pan-
European service or will there be a role for intermediaries to
help facilitate deals with a number of operators depending
on network coverage?56
• Will there be scope for joint ventures/co-investment models
or risk sharing between the telecoms and the automotive
sector to share the burden of network deployment
(particularly in costly deployment areas such as rural areas)?
• Who will be expected to pay for the connectivity - for
example will OEMs bundle in the price or will they expect
consumers to contribute to the connection costs through a
subscription?
4.2 Media and entertainment

Consumption of media and entertainment content on the move
and the increase in user-generated content has led to increased
demands on mobile data networks in recent years. High quality and
high-resolution audio-visual services will continue to drive
increasing demand for higher downlink data rates, whereas user
generated content such as sharing of live-video over social media
are driving demand for increased uplink data rates. For example,
according to the latest Ericsson mobility report, “[m]obile video
traffic is forecast to grow by around 50 percent annually through 2022
to account for nearly 3 quarters of all mobile data traffic. Social
networking is expected to grow by 38 percent annually over the next 6
years.”57
56
For example, where car manufacturers are reliant on cellular networks for
connectivity opportunities may arise for new intermediaries who help car
manufacturers to negotiate deals with multiple operators across multiple regions
to ensure high coverage across countries and boarders, providing a ‘one-stop-
shop’ for 5G connectivity.
57
Ericsson, “Mobility Report”, June 2017. Available at:
https://www.ericsson.com/assets/local/mobility-report/documents/2017/ericsson-
mobility-report-june-2017.pdf
35
Use cases for 5G
To the extent that media and entertainment will continue to drive
the significant increases in mobile data usage, a key challenge will
be to provide ways of supporting this growth and guarantee a high
level of quality. This might be particularly relevant at large public
events or stadia where video content is being consumed or created.
5G may help support these requirements in a cost-effective manner.
It might also allow of new, data hungry services. For example,
spectators at events might be encouraged to enhance their viewing
experience through virtual reality headsets, augmented reality
services, or by allowing spectators to view video replays and live
video streams from different camera angles on their mobile device.
5G’s role in supporting high-quality streaming is not limited to
supporting high data rates in dense areas; it might also have a role
to play in the development of broadcast and media distribution
platforms. For example, as consumer habits change from watching
linear TV on a stationary set at home, we have already seen an
appetite for catch-up TV and a greater number of devices used for
viewing such services. This has required broadcasters to rely on
various forms of local caching to distribute content cost effectively
(for example, through the use of commercial content distribution
networks).
There will be increased need for the seamless integration of
different network technologies, so consumers can move between
them and consume video content without disruption. Whilst at
present, distribution networks might largely be considered separate
and independent (for example, distribution over cellular, satellite,
broadcast airwaves, digital terrestrial television etc.), 5G may
facilitate the development and provision of a single solution that
can “exploit delivery modes for unicast, multicast, broadcast as well as
local caching.”58
Whilst there are some uncertainties in how future media production
and broadcast will evolve, 5G networks may bring advantages to
the broadcasting sector that help remove production from resource
heavy, on-site, outside-broadcast units to a central location where
video streams gathered from multiple cameras to be compiled.
Whilst this may, in part, rely on fixed infrastructure, 5G might help
where infrastructure is lacking, or where wireless connectivity
provides a more cost-effective solution.
Potential use cases
Table 3 below gives some examples of use cases for the media and
entertainment sector, the expected impact and the critical
requirements for connectivity.
58
5G-Xcast website. Available at: http://5g-xcast.eu/about/
36
Use cases for 5G
Table 3: Potential use cases in the media and entertainment industry

Supporting high quality Will allow users to engage Data throughput > 15 Mbps
data services in dense with media services at
areas for improved large events and may Connection density > 150000 /km2
viewing experiences and support the emergence of Latency < 10 ms
sharing of user and new, data hungry services
machine generated which spectators at
content events will be encouraged
to use to enhance their
viewing experience
Immersive media Support the emergence Data throughput > 30 Mbps

of interactive media and
improved, immersive Latency < 10 ms
video
New distribution Allows for a shift away Data throughput > 10 Mbps
technologies from traditional
distribution towards IP Reliability > 99 %
and to explore the use of Latency < 10 ms
multicast and broadcast
modes over wireless
networks
Cooperative/off-site Helps to remove Data throughput > 15 Mbps

media production production from resource
heavy, on-site, outside- Reliability > 99.9999 %
broadcast units to a Latency < 10 ms
central location
potentially bringing cost
savings and more efficient
use of resources
Enhanced capabilities of 5G that could support these use cases

For provision of high quality mobile data services in dense areas
(such as events or stadia), 4G has proved to be significantly better
than 3G. However, if data requirements continue to increase,
especially when we might expect to see the emergence of new,
data hungry services (for example, virtual reality headsets,
augmented reality services, live replays or video streams from
different cameras at an event) 5G’s ability to accommodate high
data rates and a high density of devices might be valuable.
Furthermore, broadcasters want to explore new distribution and
production methods taking advantage of transfer of information
over IP (rather than traditional broadcasting technologies). Fixed
Ethernet infrastructure may provide some solutions. However, 5G
might allow for a much greater reliance on wireless technologies
and facilitate the migration of media content and services from
37
Use cases for 5G
legacy systems by facilitating improvements in the physical,

transport and application layers of the radio network.59
In particular, for broadcasters 5G may help with delivery of
broadcast over cellular networks at the radio level within the core.
Whilst broadcast mode exists for 4G it is time sliced with unicast
traffic. We understand that in 3G-PPP, under Release 14, there was
work to support a standalone broadcast mode.60 5G may allow for
the possibility of transmitting broadcast and unicast data
simultaneously which could have efficiency benefits, however the
technical work to determine this possibility is still in the early stages.
In any case, broadcasters (especially those public service
broadcasters under a universal service obligation) must ensure that
their content remains available to all, so at least in the short-to-
medium term where 5G availability might be limited to certain
areas, other available broadcast technologies must continue to be
available.
Expected role of the media and entertainment industry in driving
5G deployment
5G’s ability to accommodate high data rates will be particularly
valuable in the media and entertainment sector. In the short-term
we consider that the desire for improved data rates for
consumption and generation of media and entertainment, coupled
with the potential for the emergence of even more data-intensive
services could be one of the key drivers behind the ‘advanced
mobile broadband’ use-case for 5G. According to the latest Ericsson
‘5G Readiness Survey’61 “Media & Entertainment is the top industry
sector that operators are focusing their 5G activities on” and “[h]igh-
quality streaming to mobile devices stood out as the single most
important use case in the Media and Entertainment sector.”
We explore some example use cases in more detail in Annex C with
a focus on enhanced mobile broadband to support on-site live
event experiences with significant data rate and device density
demands. Annex C also considers broadcast and multicast
communication enablers for 5G, based on the lessons from the so-
called “5G-Xcast” project.
To the extent that the media and entertainment industry
requirements are supported by eMBB, the key issues are around
59
5G-Xcast website. Available at: http://5g-xcast.eu/about/#concept.
60
For example, we understand that that advances in 3GPP Release 14 will include a
standardised interface between mobile network operators and service providers
and to the 3GPP system itself – for media delivery and control, radio enhancements
for improved broadcast support and system enhancements to allow delivery of
free-to-air receive-only services.
61
the road to 5G””.
38
Use cases for 5G
how networks are deployed to meet the significant demands that

might be created by a mass-market service using high data rates.
This is likely to create a need for network densification in urban
areas or areas with high capacity demands (such as stadia). In the
longer run, the use of high frequency mmWave bands and the
installation of small cells will be crucial for meeting these capacity
requirements.
4.3 Manufacturing
In 2012, Europe set a target of a GDP contribution of 20% from the
manufacturing sector but is some way from realising this62.
Therefore, in recent years, the European manufacturing sector has
been trying to use ICT enhancements to improve productivity. The
digitisation of manufacturing is sometimes referred to as the ‘fourth
industrial revolution’ and is expected to be fuelled by cyber-
physical-systems (CPS) and the Internet of Things (IoT) to enable
effective, connected and flexible factories of the future.
5G could contribute to further developments in the manufacturing
sector by enabling improvements in wireless data rates through the
evolution of mobile broadband networks. For example, there is
scope for the sector to take advantage of improvements in
connection density and lower latency as it will enable multiple
devices and machines to communicate. This could facilitate new
ways of working, bring significant efficiency gains and disrupt
traditional manufacturing methods.
The growing importance of CPS production has created a demand
for increased connectivity and information exchange over a larger,
more diverse set of devices, at new scales and requiring fast
response times. 5G could help the manufacturing sector to realise
the concepts of the connected factory and connected products, and
to participate in connected value chains.
Potential use cases
A number of potential use cases for 5G have been identified63 within
the industrial sector, covering both the manufacturing processes
within the walls of a single factory and integrating processes across
different factories. We present the most widely reported use cases
62
The European House – Ambrosetti, Italy - ¨European manufacturing between
structural trends and future perspectives: Taking up the challenge¨, Industrial
policies for global manufacturing, World Manufacturing Forum, 2014.
63
Based on 5G-PPP’s White Paper: 5G and the factories of the future, 2015.
39
Use cases for 5G
in Table 4 below. Many of these fall under the larger umbrella of

mMTC or the IoT.
Table 4: Potential use cases in the manufacturing industry
Cell automation: devices Flexible and highly Latency < 0.5 ms

in an assembly line and efficient production
control units Reliability >99.99999%
communicate wirelessly
Automated guided Increased safety, Mobility > 10 m/s

vehicle: autonomous efficiency
vehicles to transfer goods Reliability > 99.999%
in a factory
Process automation: a Increased efficiency, Reliability >99.99999%

high number of low flexibility, lower inventory
maintenance sensors and
actuators communicate
wirelessly with control
units
Logistics tracking: track Increased efficiency (cost Connection density > 1000000 /km2
flow of goods from raw and time)
material to delivery
Remote assistance and Increased Reliability > 99.999%

robot control: remote product/process quality
control of robot to fulfil
operations such as
measurements, digging
Augmented reality (AR): Increased efficiency, Data throughput > 500 Mbps
live direct or indirect view worker satisfaction, safety
of a physical environment
for training and
maintenance

We consider, in the short-to-medium term (the timeline over which
we are concerned in this study) the requirements for a number of
these use cases primarily arise from an increasing number of IoT or
M2M-enabled devices. To some extent, these requirements for
wireless connectivity within an enclosed environment (for example
a factory) may be met using existing technologies (such as NB-IoT,
Lora, SigFox etc.) possibly complemented by RLAN solutions such as
Wi-Fi. Wi-Fi networks are continuing to evolve, and advances may
be sufficient to support some applications. For example, work is
40
Use cases for 5G
underway to allow for “a more intelligent wireless network [that] will

be introduced by bringing 3G and 4G cellular concepts to Wi-Fi” 64.
There are possibilities for the standard to allow integration with 5G
networks (e.g. the IEEE 802.11 ax Wi-Fi standard). Wi-Fi Certified
WiGig allows for multi-gigabit, low latency connectivity.65 The Wi-Fi
Alliance is introducing Wi-Fi power saving features to enable use for
IoT scenarios with battery-powered devices.66
However, if the number of connected devices grows strongly and
the connectivity requirements of those devices increase with
respect to data throughput or low latency, it is possible that existing
technologies may fail to meet these requirements cost effectively.
5G may become a superior alternative at some point. For example,
where IoT devices must be cost and energy efficient, ultra-reliable
and deliver very low latencies, while also supporting a dynamically
configured factory floor, 5G has the potential to provide power
saving techniques to reach 10 to 15 years battery lives, and more
flexibility in terms of target performances (with reliability of up to
99.9999%) and ultra-low latency. This could ensure the same level of
reliability as offered in current wired architectures.
5G may also be attractive for manufacturers who would otherwise
need to ensure that diverse wireless IoT systems work well together.
To manage coexistence of wireless technologies and assure
interoperability between communication systems, either protocols
are required to manage the cooperation of technologies working in
the same frequency bands (which may be risky) or usage must be
spread over multiple frequency bands. In contrast, self-organizing
5G technology can allow a large number of sensors to be operated
in close proximity without compromising availability and without
needing complex planning.
Furthermore, machine interactions might still be executed using
wired technologies if there is a lack of technology to orchestrate,
monitor and optimise wireless networking. Such orchestration
might be difficult or costly if manufacturers adopt a diverse set of
wireless technologies, protocols and data formats. In contrast, using
5G for all of a manufacturer’s connectivity requirements would
meet this need; furthermore, management capabilities can be
extended beyond networking aspects to include services for
security, data analytics and cloud/edge computing.
64
https://blogs.cisco.com/wireless/a-glimpse-into-your-future-802-11ax
65
https://www.wi-fi.org/discover-wi-fi/wi-fi-certified-wigig
66
https://www.wi-fi.org/news-events/newsroom/wi-fi-alliance-introduces-wi-fi-
power-saving-features
41
Use cases for 5G
Expected role of manufacturing industry in driving 5G

development
The sheer diversity of manufacturing use cases, as outlined above,
represents a challenge as well as an opportunity. The diversity of
manufacturing use cases creates a need for information sharing in
order for providers to understand better what particular sectors
might require. However, manufacturers are taking an interest in the
development of 5G and active partnerships are being formed
between industry and ICT players, such as the I4MS (ICT Innovation
for Manufacturing SMEs) of the European Commission, to drive
innovative technology adoption in manufacturing. Thus, the
manufacturing sector is one where the stakeholders involved, from
the industry players to equipment manufacturers and operators, are
likely to consider the ways in which 5G deployment can help
support the industry. However, questions remain about who might
provide connectivity in controlled private spaces such as factories
and how this might be charged for.
4.4 Logistics
Logistics and Supply Chain Management (SCM) are concerned with
the flow of products and information between various points of the
distribution chain – typically from manufacturers to distribution
centres, from distribution centres to retailers and from
manufacturers or retailers to consumers. IoT is expected to have a
great impact in the way companies approach logistics and SCM.
Technology is currently having a significant impact on the logistics
sector. The key trends are shown in Table 5 below. These are likely
to have a substantial impact on the industry in the long term and
create opportunities for 5G to be used. Even though most of the
upcoming requirements of the logistics sector can be satisfied by
the existing technologies, 5G may have a role in the sector in the
future by providing more efficient fleet management and delivery
solutions.
42
Use cases for 5G
Table 5: 5G Opportunities via convergence of technology and logistics trends
5G
Technology trends Impact on Logistics
Opportunity
Improved supply chain transparency, safety, efficiency, and

High speed Internet High
efficient resource planning
IT standards Horizontal collaboration, more efficiency and transparency Low
Improved customer experience, inventory management, and

Data analytics Medium
predictive maintenance
Cloud New platform-based business models, increased efficiency Medium
Blockchain Supply chain security, reduction in bottlenecks and errors Low
Robotics & automation Improved delivery and warehouse management, lower costs High
Autonomous vehicles Increased delivery efficiency Medium
UAVs (Drones) Cost and delivery efficiency High
3D Printing Transportation efficiency Low
Potential use cases

The logistics industry was among the first adopters of the IoT within
operations, from the introduction of handheld scanners to sensors
that monitor cargo integrity and transportation performance.
However, this is just the start, with many more specific use cases
expected, as shown in Table 6 below.
43
Use cases for 5G
Table 6: Potential use cases in the logistics industry
IoT driven smart inventory Enhance warehouse Connection density > 1000000 /km2
management: Use of efficiency, ensure service
pallet tagging, cameras, quality Battery life > 10 years
sensors, automation to
achieve damage
detection, real time
visibility and accurate
inventory control
Optimal asset utilisation: Reduce errors, increased Connection density > 1000000 /km2
Connecting machinery workforce efficiency
and vehicles to a central Mobility > 100 km/h
system to monitor all
assets in real time and to
enable predictive
maintenance.
Fleet and asset Optimised routes, Connection density > 1000000 /km2
management: Use of improved fuel economy,
sensors and wireless reduce deadhead miles Battery life > 10 years
connectivity to create
fleet efficiencies
UAVs or Drone delivery: High speed delivery, Reliability > 99.999%

Use of drones to deliver access to low population
packages by using a zones Latency < 5 ms
remote-control system
Truck platooning: Increase capacity and cost Latency < 5 ms

Platooning of delivery efficiency
trucks equipped with Reliability > 99.99999%
mutually connecting
smart technology.
Connected Ports: Devices, Increase in trade, accurate Connection density > 200000 /km2
machines and humans tracking, secure ports
sharing real time Battery life > 10 years
information to enable
ports to work in a smarter
way

The demand for real time information and delivery predictions has
put the ICT sector at the centre of the logistics industry. Highly
automated systems are required to meet the demands of the
industry. Existing technologies already have the potential to enable
many of the use cases in Table 6, but 5G is likely to offer certain
enhancements. In particular, IoT applications in large warehouses
and logistics hubs might have high device densities, with a need for
lower power sensors to tag items; 5G would be able to meet these
requirements. There may also be some need for high reliability
connections for real-time control of automation such as drone
deliveries.
44
Use cases for 5G
Expected role of the logistics industry in driving 5G development

The logistics industry could be an early adopter of 5G as it becomes
more widespread. However, the industry is not likely to drive the
development of 5G because, as previously stated, most of the use
cases can be realised with the existing network technologies. There
is significant push, however, on use cases such as drone delivery,
truck platooning, smart ports, etc. through partnerships with
adjacent players who are relevant stakeholders in the value chain.
Amazon and Google are working on drone delivery projects, and
automotive players such as Scania are driving projects on truck
platooning.
The key requirement of the logistics sector is the need for
connectivity ‘on-the-go’. Therefore, the key questions are whether
5G applications within the logistics sector will just rely on public 5G
networks or rather fall-back to use other forms of connectivity such
as 4G.
Real-time control of delivery drones would require extensive 5G
networks, covering the entire area of operation. Therefore, such
applications are unlikely within the near term and, in the longer
term, potentially limited to areas such as cities with large swathes of
contiguous 5G coverage. Even then, there is a question about how
the reliability and security of such services would be assessed given
the possibility of significant harm to the public if they failed. Larger
area 5G roll-out would be necessary for widespread use. In the
absence of low latency 5G connectivity allowing real-time control, it
might be possible to command drones over less capable networks
by relying to a greater degree on autonomous behaviour. In this
sense, many of the key issues are very similar to those in the
automotive sector. The pace of such developments is likely to be
highly dependent on how specific sectoral regulations (e.g. safety
rules for drones) evolve.
4.5 Agriculture
Agriculture is rapidly adopting technology, as it evolves from an
industry entrenched in tradition to one that is fast embracing
change. Technological innovations are automating labour-intensive
tasks and providing farmers with greater knowledge and insight
into their crops and environmental factors that can be used to
increase efficiency and yield. The agricultural sector has already
started to embrace ICT and wireless connectivity to increase
productivity and efficiency. IoT can help achieve smart farming and
‘precision agriculture’ that will be vital to achieve improved crop
yields. For example, according to the UN Food and Agriculture
Organisation, farmers across the globe will have to grow 70% more
food in 2050 than they did in 2006, so technology’s ability to
support increases in efficiency could be particularly valuable. Data
45
Use cases for 5G
collected from agricultural machinery (e.g. combined harvesters)

can be wirelessly backhauled to allow for detailed analysis, creating
location-specific plans for dosing of fertilisers that can be uploaded
to tractors. IoT can also be used to keep track of livestock and
monitor farm assets.
To the extent that precision agriculture can reduce use of fertilisers
and pesticides, applying only where necessary, this could have
significant environmental benefits. For example, run-off into water
courses from excessive use of nitrogen fertilisers could be reduced.
The global trends in the industry suggest that the opportunities for
5G include:
• Expansion of smartphone and internet penetration:
Farmers are becoming increasingly reliant on smartphones
and other intelligent mediums to stay updated and
participate in industry-specific knowledge sharing
initiatives. Smartphones and tablets are likely to be
important in accessing analytical and other services for
precision agriculture. 5G could provide improved
capabilities to meet the growing demand of broadband
connectivity for the farmer community, where alternatives
are not already available or cost effective. 5G is likely to be
used as the technology for future FWA deployments.
• Increasing adoption of technologies: Farmers, as well as
the farming processes, are increasingly becoming
dependent on technology. Low power wide area networks
using Zigbee or LoRa, Wi-Fi and various other wireless
sensor technologies enable farmers to collect data to
streamline numerous agricultural operations, such as
purchasing, inventory control, planting, monitoring, and
harvesting. 5G can provide improved capabilities,
bandwidth and provide a platform for advanced processes
such as precision farming, drone-based crop monitoring etc.
which could prove valuable and cost efficient to the
farmers.
However, the extent to which the agricultural sector (located mainly
in rural areas) could use 5G will of course depend on penetration in
those areas.
Potential use cases
Table 7 summarises the main use cases that are emerging in the
agriculture sector. They are largely applications requiring networks
of sensors. However, there is also the potential for the use of drones
for remote sensing, which raises similar issues to the use of drone
from deliveries. Such applications either need low latency
connections to allow real-time closed-loop control (for which 5G
would be appropriate), or else drones need to be autonomous to
some degree, using a higher latency network for command data. In
46
Use cases for 5G
either case, safety issues and sectoral (i.e. aviation) regulation are
likely to constraint the speed of adoption of these new
technologies.
Table 7: Potential use cases in the agriculture industry
Precision farming: Use of Ensure profitability and Connection density > 100 /km2
sensor data to measure sustainability, protect
crop yields, moisture environment
levels and terrain
topography
Smart Irrigation: Use of IoT Higher irrigation Connection density > 100 /km2
to measure humidity, soil efficiency
moisture, temperature Battery life > 1 year
etc. to calculate precise
requirements for water
Agriculture drones: Use of Enhanced protection, Reliability > 99.999%

UAVs to monitor crop Efficient inspection and
health, scan areas, monitoring Latency < 5 ms
agriculture photography
etc.
Soil and Crop monitoring: Enable informed farming Connection density > 100 /km2
Use of sensors to monitor decisions, minimise
moisture and identify erosion Battery life > 1 year
issues such as diseases or
insects
Precision livestock Ensure optimal yield, Connection density > 100 /km2
farming: Real-time enable informed farming
monitoring of decisions Battery life > 1 year
productions, health, and
welfare of livestock

In order to realise smart farming and monitoring, IoT platforms
require a wireless internet connection that is both fast and reliable.
Although it is true that 5G could bring enhanced capabilities in
terms of fast and reliable connections, current technologies, which
are continuously evolving, have the capabilities to enable most use
cases expected in the agriculture industry. For instance, farmers can
already implement IoT-based precision farming solution comprised
of sensors and gateways using existing NB-IoT standard
technologies or long-range communication technologies such as
LoRa, to meet the demands in a cost-effective manner.
There are various ways of creating narrowband sensor networks
without needing 5G (or even any cellular connectivity). For example,
suppose that a large farm might want to remotely monitor sensors
that can detect soil moisture, crop growth and livestock feed levels,
47
Use cases for 5G
so that actions such as watering or applying crop fertiliser can be

optimised. This could be done by deployment of a number of ‘slave’
devices connected to a master ‘hub’ using low power wide area
networks such as LoRa, with only that hub device being connected
to the mobile network using a normal SIM to 4G or 3G network or
even a fixed connection. The hub then sends the data to servers to
be analysed and interpreted. In areas with very challenging
connectivity, backhaul of a hub might even use a satellite
connection. Clearly such models become increasingly unattractive
as the amounts of data flowing from sensors become greater; for
example, the widespread use of high quality video feeds would
probably not be feasible on that basis.
It is expected that equipment and machinery in farming will
become autonomous moving forward. Real-time control of such
equipment over a wireless network would require a low latency
connection and ultra-secure connectivity and control platforms.
Thus, even though currently the existing technologies are sufficient
to cater to the needs of the industry, in the long-term - beyond the
timeframe of this study - some agricultural use cases could become
dependent on 5G. For example, agriculture drones that monitor
crop health, scan areas, agriculture photography and so on might
benefit from the capabilities of 5G.
We understand that operators such as Vodafone are already
working with partners on the use case of automatic drones for
precision farming. We expect 5G-based use cases such as
agricultural drones, despite considerable current interest, to be
realised beyond the timeframe of 2018-2022.
Expected role of agriculture in driving 5G development
Within the timeframe of this study, we would not expect the
agriculture industry to be a key driver of 5G deployments, mainly
due to the sufficiency of existing networks to fulfil the required
needs for low data rate sensing, and also potential coverage issues
in the short-to-medium term, with 5G deployments likely to be
focussed on urban areas initially. We expect markets for various
agricultural support services based around collection and analysis
of data to develop on the basis of a mixed connectivity model (i.e.
various fall backs to less capable networks as discussed above)
rather than 5G availability being a prerequisite for these
developments.
Nevertheless, as we discuss in the latter sections of this report, there
may be some examples where 5G, if priced attractively, could
provide a more cost effective and flexible solution to providing
connectivity for agricultural applications in certain cases.
For example, for dense sensor networks, existing solutions can, in
some cases, have some disadvantages that could be addressed by
5G. For example, using a spoke-hub or master-slave configuration
might require professional design/configuration of the topology or
at the very least significant configuration by the installer.
48
Use cases for 5G
Furthermore, as the number of devices increases there may be

some problems regarding device density, since increasing the
number of devices hinders the topology design, organisation of
communication of all devices. In contrast, deployment of dense
sensor network – even if data rates are modest – might be much
easier using a public 5G network. The pricing and charging structure
of the 5G ‘solution’ will be an important factor in generating
demand for using 5G networks for such purposes, especially where
the incremental benefit of 5G may be limited (i.e. not essential for
the provision of the service).
4.6 Energy and utilities

Energy systems are currently facing the following challenges:
• Increased consumption of electricity creates challenges for
management of the grid, especially where demand is
unpredictable. This may be made worse in the short-term
by increasing electrification of transport, leading to an
increased demand on energy consumption at potentially
unpredictable locations and times;
• Variable generation technologies - renewable energy
generation such as wind turbines whose output is not
controllable - make up an increasing proportion of the
overall supply due to decarbonisation, creating a growing
need to manage the transmission network flexibly. This
includes being able to control energy storage facilities such
as batteries.
In addition, issues with the current infrastructure are emerging,
which suggests they may not be fit for purpose. Increasing
demands are likely on transmission grids given that new renewable
generation capacity is often far from the centres of most demand.
Where infrastructure is aging and possibly more susceptible to
faults, which will need to be identified quickly. Where systems have
been dimensioned to meet peak demand, such that during non-
peak times the system may be underutilised, there is a clear role for
improved capacity management to make the most of such
redundancies as and when required. Overall, this potentially
increases the need for monitoring and control within supply grids.
Some new transmission infrastructure is also likely to be needed in
many EU Member States to respond to these changes in both
demand and supply.
5G could help provide the necessary support for critical machine
type communication (MTC) applications for energy grid monitoring
and control and support the massive volume of MTC type
applications emerging from increased use of smart metering.
According to the 5G-IA: “the anticipated performance and flexibility of
49
Use cases for 5G
5G will enable a communication infrastructure which is able to support

the emerging energy use-cases…the ongoing evolution of the power
grid into a grid supporting a much more distributed generation and
storage of power as well as micro-grids would be a clear beneficiary of
the high performance, but still very flexible communication
architecture provided by 5G.” 67
Potential use cases
A number of ‘smart’ or ‘connected’ use cases for 5G have been put
forward to try and address some of these issues in the electricity
sector, with the main focus being on ‘smart meters’ and ‘smart
grids’.
‘Smart metering’ could be used to provide customers with greater
information about their energy use, in near real time, allowing them
to make more efficient decisions. This could help smooth peak
demands, especially if consumers can respond to real-time market
price information. Furthermore, by informing the supplier of exactly
the energy they use, removing the need for manual meter readings,
customers will get more accurate bills and suppliers can better
manage activities and investments based on the improved
information.
‘Smart grids’ refer to everything in the grid being connected,
monitored and controllable. They have been proposed as a way of
better managing the energy grid, to react to the challenges of
unpredictable energy generation (e.g. from renewal technologies
which may begin to replace “dispatchable and controllable base load
generation”68) and as demand for energy and patterns of use evolve
as well. For smart grids, there will be a need for an ultra-reliable and
secure network to connect power generation, transmission and
distribution assets and management systems. Even where grid or
distribution operators have fixed (data) network connections to
facilities for monitoring and control, in some cases they may need a
second backup connection for reliability.
Enhanced capabilities of 5G that could support these use-cases?
Whilst ‘smart metering’ could be considered as part of the IoT
category and mMTC applications, such services are already available
and in use and functioning without the need for 5G given their
relatively low data volumes and no requirements for extremely low
67
5G Infrastructure Association, “5G and Energy – version 1.0”, 30 September 2015.
68
50
Use cases for 5G
latency.69 Smart metering is not mission-critical – unlike monitoring

and control of the power grid – and can use shared spectrum.
Therefore, we do not expect smart metering to be a key use case for
5G.
For smart grids, a large number of connections within the energy
grid are already in place, often relying on Local Access network
Solutions based on Ethernet infrastructure.70 However, as the grid is
required to become ever ‘smarter’ there will be an increasing need
for communication between sensors, control systems and energy
generation and storage transmission assets 71. As more and more
elements of the energy grid need to be connected and controlled,
the demands on the communication networks to support this will
increase. These connections need to be highly reliable and secure.
There is currently considerable concern about the possibility of
cyberattacks on key national infrastructure, such as power grids and
other utilities.
The 5G-IA, considers that for those assets in a smart grid network
that currently have no communications or measurement
equipment 5G could provide an economically viable wireless
solution compared to full-fibre fixed connections but with the
added advantage of improved resilience compared with existing
mobile technologies.72 This view is supported by Vodafone that
consider “5G networks can be used for monitoring and control of the
grid in places where fibre networks have not been rolled out yet or
where this would be too costly (e.g. rural areas). The resilience and
reliability of 5G will give utility providers the confidence to push the
technology deeper into the generation and distribution networks. It is
also expected that the benefits of more control deeper down the
network will trigger efficiency gains.”73
69
However, the 5G-IA considers there application might have greater requirements
in the future: “They will reach near real time application in coming years, enabling
near real time optimisations of sections of the low and medium voltage infrastructure
with impacts on the communications requirements of utilities towards the customers
(residential and business) particularly in urban areas where 5G will become available
and can address the communication needs of smart meters if the design targets are
accordingly set.” 5G Infrastructure Association, “5G and Energy – version 1.0”, 30
September 2015.
70
The 5G IA gives a more complete overview of the various smart grid
communication domains and the types of networking they rely on. See 5G
Infrastructure Association, “5G and Energy – version 1.0”, 30 September 2015.
71
Huawei, “5G Opening up New Business Opportunities”, August 2016. Available at:
http://www.huawei.com/minisite/hwmbbf16/insights/5g_opening_up_new_busin
ess_opportunities_en.pdf
72
73
Vodafone: https://www.vodafone.com/content/dam/vodafone-images/public-
policy/reports/pdf/gigabit-society-5g-14032017.pdf
51
Use cases for 5G
Expected role of the energy and utilities industry in driving 5G

development
The question is when these requirements will be realised and
whether we can reasonably expect 5G to be used in the energy and
utilities sectors before 2022 (the time-period covered in this study).
Although there does not appear to be a coordinated approach or
roadmap for the emergence of ‘smart-grids’ in Europe there are
roadmaps for some countries. For example, the Sustainable Energy
Authority of Ireland has set out a roadmap to explore how a smart
grid could be operational in Ireland by 2050, part of which includes
the establishment of a 5G test bed facility in the run up to 2020 and
engagement with large scale technology R&D projects from 2020
onwards.74 This suggests that the emergence of smart grids reliant
on 5G may still be some time away.
According to Ericsson’s ‘5G Readiness Survey’, which asked
telecoms operators about the industries and use-cases they were
considering and pursuing, less than a third of those surveyed
included “Energy and Utilities” within their top three industry sectors
being pursued.
Furthermore, as highlighted by Vodafone, there are some additional
challenges to be overcome in this industry, which may limit the
scope for the emergence of 5G-use cases within the timeframe of
the study (or at least suggest that energy and utilities will certainly
not be a driving force behind 5G deployment). For example, the
industry may be slow to adopt due to the fact that “the energy
industry is generally considered as being risk averse and conservative,
meaning 5G will have to be stable and technology proven before take-
up” 75. A further challenge arising in this sector due to the integrated
grid system is the focus on security of the power system, which
would require any new technology to meet the security standards.
Grid operators can be expected to have a preference for deploying
their own wireless networks (for example using LTE-M in the 410-
420 MHz range) given concerns about whether mobile operators
public networks will fully meet their needs.76 A move towards
reliance on external 5G networks may require a significant cost
advantage to be demonstrated to outweigh these concerns.
74
See Sustainable Energy Authority of Ireland, “Smartgrid Roadmap 2011-2050”.
Available at: https://www.seai.ie/resources/publications/Smartgrid-Roadmap.pdf
75
Vodafone: https://www.vodafone.com/content/dam/vodafone-images/public-
policy/reports/pdf/gigabit-society-5g-14032017.pdf
76
For example, see ESB’s response to ComReg’s consultation on the proposed
release of the 410-415.5/420-425.5 MHz subband, December 2017, available at
https://www.comreg.ie/publication-download/non-confidential-submissions-
comreg-document-1767-proposed-release-410-415-5420-425-5-mhz-sub-band
52
Use cases for 5G
Therefore, although there may be some prospects for 5G uses in the

energy sector, they may not be forthcoming, especially given that in
the short-medium term, it may be that 5G finds itself competing
with existing wired/fibre solutions or with self-supplied wireless
solutions with regards to reliability, performance, cost and usability.
There may be some situations where 5G becomes an attractive
option for connectivity within the grid, however, we anticipate that
the utilities sector may be slow to transition and that it will be very
unlikely to be a key driver being 5G deployment. Rather, utilities are
more likely to look at 5G as an alternative once infrastructure is in
place and reliability has been proven.
4.7 Healthcare
Health systems are faced with significant challenges from
constantly increasing healthcare demand due to growing and
ageing populations. Cost of healthcare as a percentage of GDP
continues to grow faster than average economic growth.77 New
opportunities to improve the cost efficiency of healthcare delivery
are extremely valuable to the industry.
Opportunities for increasing efficiency and reducing costs will arise
from applications that allow for the collection and management of
more data and opportunities such as improved health monitoring,
smart medicine and remote diagnosis and surgery. Big data
management and sharing could be a key driver of more informed
medical decisions and help prevent illnesses.78 The use of remote
healthcare (“Telehealth”) could also significantly reduce costs by
enabling the decentralisation of medical care, which will be
particularly advantageous in rural or hazardous areas.
It is likely that these applications will rely on a range of technologies
depending on the specific requirements. For example, some could
rely on fixed connections whilst others might be able to rely on low
data rate wireless technologies (2G or 3G) or even over Wi-Fi for
example. It is possible that some use-cases might also see potential
in 5G.
Healthcare providers believe 5G has a role to play in bringing
improvements and significant benefits to the sector. For example, in
a study by Ericsson, almost three quarters of surveyed healthcare
77
5G-PPP, White paper on 5G and e-Health, September 2015.
78
Tech4i2, Real Wireless, Trinity College Dublin, InterDigital, “Identification and
quantification of key socio-economic data to support strategic planning for the
introduction of 5G in Europe, A study prepared for the European Commission”, May
2015.
53
Use cases for 5G
executives expect 5G to enable new services and products that will

improve quality of life.79 Implementing 5G technologies is expected
to enable digitalisation and virtualisation of healthcare, bringing
benefits through reducing costs of healthcare provision and
through greater access to patient information.
Potential use cases
Table 8 below presents examples of use cases for the healthcare
sector,80 the expected impact within the sector and critical
requirements for connectivity.
Table 8: Potential use cases in the health sector
Remote monitoring of Could provide new Battery life > 15 years

health or wellness data opportunities for
through wireless devices monitoring well-being Connection density -
remotely and for Coverage (indoors) -
providing better medical
information to healthcare
providers, including
medical data/information.
Smarter medication Linked with health Battery life > 15 years

monitoring devices,
medication could be Connection density >1000/km2
administered immediately Coverage (indoors) -
when required.
Wireless tele-surgery May allow specialists to Latency < 1 ms

join a local surgeon
remotely, or allow for Reliability > 99.999%
surgery to take place for Data throughput > 10 Mbps
those patients in remote
or dangerous locations. Coverage -
Security -
Assets tracking and Cost savings and Connection density >1000/km2

management in Hospitals efficiency benefits
through improved Location accuracy < cm
tracking of valuable assets Seamless handover -
within the hospital and between different access
management of technologies throughout
pharmaceuticals e.g. the hospital
tracking those coming to
end of expiry data and/or
having automated
ordering systems in place
to re-stock key items.
79
Ericsson, “Opportunities in 5G: The View from Eight Industries”.
80
Based on information presented in the 5G-PPP, White paper on 5G and e-Health,
September 2015.
54
Use cases for 5G
The use of 5G to support wireless robotic assisted surgery is widely

expected to bring a major breakthrough in tele-surgery, which
would create requirements for data speed, latency, reliability and
security81. In particular ultra-low latency will be crucial to ensure
there is zero perceived delay.82 However, many remote surgery
cases would likely rely on fixed networks with wireless 5G being the
exception for very particular circumstances (discussed below).
One eHealth application that stands to drastically increase ‘big
health’ data flows is ‘smart wearables’. Smart wearables will consist
of multiple low power, waterproof devices and sensors that can be
integrated into clothing. These sensors might collect data on a
number of environmental and health attributes such as
atmospheric pressure, temperature, blood pressure, blood glucose,
etc. 83 With the introduction of wearables, healthcare providers will
be able to collect data from more locations and larger populations.
Monitoring of biological data has potential to be broadly beneficial.
While monitoring patients can help to avoid non-adherence to
medical treatment, the capture and analysis of data will benefit
healthy consumers by helping promote a healthy lifestyle and
preventing diseases. The capture and analysis of data will also help
ensure that patients get the most appropriate treatment available
through the increased tailoring of treatment to genomes possible
with better data on medical outcomes.84
Although remote diagnosis and surgery is already available today
through fixed networks, the surgeon currently has to rely on visual
feedback, as haptic feedback85 is not available over current
technologies. 5G is reported as having the capability to be the only
wireless technology able to provide the low latency and high
availability required for enabling particular features of wireless tele-
surgery, such as allowing the replacement of the doctor’s hands
81
Maria Lemas et al, “5G Case Study of Internet of Skills: Slicing the Human Senses”,
IEEE 2017; Ericsson, “From Healthcare to Homecare”; IEEE, IEEE 5G and beyond
technology roadmap white paper”.
82
Maria Lemas et al, “5G Case Study of Internet of Skills: Slicing the Human Senses”.
83
NGMN, NGMN 5G White Paper, 17th February 2015.
84
Tech4i2, Real Wireless, Trinity College Dublin, InterDigital, Identification and
introduction of 5G in Europe, A study prepared for the European Commission.
85
Haptic feedback refers to feedback that confirms a sense of touch for the action
performed. In remote surgery, it will give the surgeon a sense of touch while using
tele-surgical devices with robotic probes.
55
Use cases for 5G
with robotic probes, while providing the same sense of touch.86

Substantial delays in either visual data or haptic feedback can lead
to cyber-sickness, which may occur when there is a discrepancy
between the visual feedback and the feedback received by the
sensory system.87 Furthermore, where fixed connections are not
feasible (e.g. ambulances, hazardous areas, etc.) 5G would be the
only technology capable of enabling wireless remote surgery.88
For remote monitoring and big data capture through widespread
use of smart wearables, 5G may bring some incremental
improvements over existing technologies. For example, whilst
bedside monitoring devices are already in place in a number of
hospitals 89 the next step is to monitor consumers outside of the
hospital. Although some smart monitoring devices can already be
delivered over existing technologies, the use of 5G for these
wearables could provide incremental enhancements. For example,
5G will be able to provide low power wide area machine type
communication and “has the ability to support a plurality of devices
and sensors, and to translate information from these devices and
sensors into critical and meaningful data points”90.
Furthermore, to protect consumer data, security is of extreme
importance in eHealth, 5G could offer assurances to safeguard
identity, privacy and authentication management for every single
device91 given 5G’s capability to support managed secure network
slices.
In cases where monitoring is life critical, these services could benefit
from 5G, as it will be able to ensure reliability and reserve the
necessary network capacity.92 However, one would expect that
there would have to be fall-back technologies to ensure that
monitoring is still possible outside of 5G coverage.
Expected role of healthcare industry in driving 5G development
A particular challenge with enabling 5G in the healthcare sector is
that, in contrast with M2M environments such as automotive and
86
(This sense of touch is received through two elements of haptic feedback, one is
tactile feedback which is touch felt on the skin such as heat and vibration, the other
is kinaesthetic feedback which is touch felt in muscle and bone such as force and
motion.)
87
Nokia, “Translating 5G use cases into viable business cases”, White Paper.
88
NGMN, NGMN 5G White Paper, 17th February 2015
89
For example, see Phillips Press Release: “Philips and VU University Medical Center
Amsterdam sign agreement on patient monitoring systems”. Available at:
https://www.philips.com/a-
w/about/news/archive/standard/news/press/2013/20130530-Philips-and-VUmc-
Amsterdam-sign-agreement-patient-monitoring-systems.html
90
David Teece, “5G Mobile: Impact on the Health Care Sector”, 26 October 2017.
91
NGMN, “NGMN 5G White Paper”, 17 February 2015.
92
Huawei, “5G – Opening up New Business Opportunities”, August 2016.
56
Use cases for 5G
industry, the applications will directly involve consumers (i.e.

patients)93. This means that before implementing any new
healthcare applications these will need to be tested and certified
and meet all of the strict regulatory requirements of the health-care
system. There will also be very strict requirements on
standardisation across any devices or technologies used (for
example, to ensure security and privacy of health information, audit
trail support and reliability). Reliance on ‘connected devices’ or
‘robots’ to administer care also raises significant issues in terms of
determining liability where things go wrong. Given these (largely
exogenous) barriers, the adoption process of wireless surgery within
the healthcare sector will likely be slow and we may not reasonably
expect to see 5G widely adopted within the healthcare industry
within the timeframe of our study. We see these as the most
significant barriers for adoption of new technologies within
healthcare and for these reasons would expect that the
introduction of 5G in this sector will be slower than in other sectors.
Of all the use-cases discussed for this sector the 5G use case for
smart wearables has scope to be realised earlier than some of the
other examples discussed as they will be used to provide additional
information to the patient and healthcare professionals without
being a critical component of the healthcare delivery process. So
safety-critical issues should be raised by smart wearables. However,
the key issues here (as with many use-cases reliant on devices with
‘embedded connectivity’) revolve around whether - despite
potential incremental improvements brought by 5G - it is attractive
relative to other technological solutions in terms of the connectivity
requirements and the extent to which the availability of alternatives
will limit what users will be prepared to pay additionally for 5G
functionality. For example, devices might log data and then
download it over a Wi-Fi network once available, as it is unlikely that
there will be a tight timing requirement. There may be few
examples where 5G is really the most cost-effective solution for
such devices which may have relatively limited demands that
absolutely require 5G.
93
Tech4i2, Real Wireless, Trinity College Dublin, InterDigital, “Identification and
introduction of 5G in Europe, A study prepared for the European Commission” May
2015.
57
Business models and value chains
5 Business models and value chains

In the previous section we considered specific use cases within a
number of vertical industries that have been widely identified as
potentially benefiting from 5G. Some of these use cases can be
considered as extensions of mobile broadband (for example the
media and entertainment cases and the infotainment in cars). Other
use cases take advantage of 5G’s distinctive technical capabilities in
terms of being able to handle a large number of devices at once
(dense sensor networks for IoT or mMTC), low latency (remote
control services) or particularly high data throughput (such as
augmented or virtual reality applications).
Whilst it is difficult to identify any single particular use case that
could, at this stage, be considered to significantly drive demand for
5G and support the investment case, it is clear that there are
potentially a number of different use cases that could be supported
by 5G.
5G’s potential to serve such a wide range of requirements relative to
that made possible by previous radio technologies, comes from the
increased flexibility 5G will allow. Network operators will be able to
configure networks in different ways to offer bespoke solutions,
which in turn opens up the prospects for telecoms operators to
provide differentiated services for a number of different verticals
simultaneously. SDN and NFV will also allow a more cost-efficient
approach to adapting networks rapidly to meet new
demands/requirements. Therefore, service innovation should
become faster and less costly; experimenting with new services
should become less risky, as little sunk investment will be needed
(at least within the network).
For this reason, the emergence of 5G could lead to significant
changes within the value chain for mobile data connectivity, both
modifying traditional business models of telecoms operators and
opening opportunities for intermediaries of various types. There is
potential for creating new “merchant markets”, where various
connectivity services are traded at a wholesale level between
operators and then different physical networks orchestrated to
create a unified ‘connectivity service’ for customers.
In the remainder of this section we consider the potential changes
to the value chain and business models that could be enabled by
5G. However, it is important to remember that, the extent to which
these will be realised in practice will depend on widespread
deployment and take up of 5G services. We consider (in Section 6)
the drivers and obstacles to the 5G business case, focusing on what
will be the immediate drivers of 5G, and potential difficulties 5G
providers may face.
58
5.1 How might 5G enable new business

models?
We have identified a number of key changes to the traditional
service provision and network deployment models that could occur
due to the emergence of 5G. Each raises potential issues that will
need to be considered in more detail by telecoms providers, service
providers and NRAs in terms of determining how best to support
the business case for 5G investment.
Service related
Differentiated The technological capabilities of 5G - including NFV/SDN and
services network slicing - open the possibilities for a larger number of
differentiated services (and various associated charging models) to
be offered. This may facilitate changes to the traditional model of
offering standardised, largely undifferentiated “mobile broadband”
services to customers in the form of a package of “minutes and
data” for a subscription fee, or charges on a per device basis based
on the data usage. For example, operators could offer different
‘services’ depending on whether customers want low latency and
high data throughput, or a service to support connectivity of
thousands of devices simultaneously.
The extent to which 5G services could be successfully marketed in
this way will depend on:
• whether MNOs (or connectivity providers more generally)
will be able to successfully identify all the relevant niches
where 5G might be useful, as these depend on the precise
requirements of the customer, the alternative feasible
technologies (especially if connectivity might be required in
areas where 5G is not available initially) and the relative
pricing of 5G and those alternatives;
• whether MNOs will be able to develop a range of
standardised tariffs to attract these various new users of 5G
(i.e. offer differentiated services and tariffs and hope
demand comes) or whether these new users may need
bespoke negotiation of pricing and service characteristics to
meet individual requirements.
Given the possibility of a number of specialist requirements, there
may be a role for intermediaries in identifying new applications for
5G and facilitating agreements between the connectivity provider
and end user (discussed below).
Therefore, there is scope for the value chain to change relative to
the status quo, with differentiated services being provided, possibly
with the help of intermediaries, as illustrated in Figure 10 below.
59
Figure 10: Value chain with service differentiation under 5G

At present: With 5G:
MNO MNO
MBB MBB
Intermediary? Intermediary?
Differentiation
Consumer/
Business Consumer
Business
Being able to differentiate services could have a number of

consequences:
• Differentiation may expand demand for mobile
connectivity. For example, a business wanting to deploy
augmented reality support on mobile in its factories may be
able to purchase a package that includes high data rates
and low latency to support this feature. In our view, this
demand expansion due to 5G is likely to come from a large
number of specialist, niche applications across many sectors
of the economy;
• Charging models need to become more complex, as pricing
will need to match the particular characteristics of the
service being offered. This might mean charging a premium
for certain services or using very different charging models;
• Differentiation by price may allow niche services to develop
and be paid for by users with specific needs, whilst avoiding
price increases for users not making use of this additional
functionality.
New opportunities As the number of connected devices increases (for example for IoT
from increased and/or M2M applications), there will be increasing demand for
demand for
‘connectivity’
individual devices to connect both to each other and to the
internet. Whilst the data requirements for an individual device may
be small, the large number of devices and, in some cases, the need
for low power requirements could be facilitated by 5G. However,
whether or not these services rely on 5G networks will rely, in part,
on the charging models for such ‘connectivity’.
At present there are no common models for charging for such
services and it would appear unlikely to be cost effective for each
device to have a separate subscription to the network. To give
appropriate incentives for different types of users, connectivity
providers (MNOs, but possibly other providers too) may have to
60
develop new charging models to provide an attractive proposition

relative to alternative technologies for many different forms of use
cases.
In addition to questions around who and how this ‘connectivity’ will
be paid for, there are also interesting questions about the different
options for how this ‘connectivity’ will be provided and the parties
involved in the service provision value chain. For example:
• The MNO may continue to provide the connection and
negotiate with individual verticals and/or equipment
manufacturers (OEMs) on a bilateral basis to provide
bespoke connectivity (option A in Figure 11 below);
• In some cases it may be that a self-supplied (i.e. private 5G
network) solution is preferred by the vertical where the
requirements are relatively well-defined by area and/or
where there may be concerns associated with relying on a
public network (option B in Figure 11 below);
• There may also be scope for joint ventures between the
vertical and/or equipment manufacturers and the network
operators (option C in Figure 11 below).
• There may be opportunities for new intermediaries to enter
the market who can negotiate deals with a large number of
mobile operators (and maybe fixed operators depending on
the specific requirements)94 and across borders and then
market a single ‘connectivity solution’ to that vertical
(option D in Figure 11 below and discussed with an example
in the box below).
94
For example, where there may be products and services that make use of
different forms of connectivity, such a 5G, 4G and WLANs. Sensor networks and the
IoT can use technologies other than 5G (indeed not even mobile networks). There
may be opportunities to create connectivity offerings that integrate these
alternatives. This is an area where mobile network operators could become active,
but there is also a need to understand the requirements of particular industrial
users, which should create opportunities for specialist services.
61
Figure 11: Business models to support embedded connectivity

(A) At present: (B) New business model 1:
Spectrum Spectrum
Self provision
using dedicated
or shared
M(V)NO M(V)NO spectrum
Vertical or Vertical or
OEM OEM
Embedded Expands Embedded

connectivity massively connectivity
End consumer End consumer
(C) New business model 2: (D) New business model 3:
Spectrum
Spectrum
MNO MNO MNO
Joint Vertical or
M(V)NO OEM
venture Fixed
operator
Intermediary
Vertical or
OEM
Expands Embedded Expands Embedded

massively connectivity massively connectivity
End consumer End consumer
Role of Option D highlights a potentially new role for intermediaries in the

intermediaries in value chain, downstream of network operators. There may be new
service provision
opportunities for intermediaries or aggregators who can offer a
single solution to those demanding connectivity, taking on the role
of the negotiator to ensure issues such as securing connectivity
across borders, roaming and billing are all taken care of.
We already see aggregators who put together such trans-national
connectivity packages for particular industries. For example, there
are aggregators offering connectivity services to car manufacturers
that remove the need to make individual deals with mobile network
operators in multiple countries (described with an example in the
box below).
62
Example: 5G Networks to support automotive applications

Where cellular connectivity is embedded, there will need to be a deal between network operators and
device manufacturers in terms of who pays for connectivity and data use.
This might require specific car manufacturers making deals with specific operators to secure
connectivity. For example, a partnership between Ford and Vodafone in Europe has been struck to
secure 4G data services within its cars.95
However, there will be requirements for such services to work across all regions and across borders.
Therefore, deals may be needed across a number of operators or a number of countries, while the costs
of network roll-out to support this use case may be particular high, especially for those areas already
falling outside 5G deployment for other services such as eMBB (for example in rural areas). In this case,
if the car manufacturer believes there is a large private value for connectivity, they may be able to fund
some of the roll-out to get coverage through co-investment models.
There is also potential for new companies to provide solutions and help facilitate deals between the
main parties. For example, one company “Cubic Telecom” has built a virtual networking solution to
allow connected vehicles to automatically connect to service providers in whichever country they are
in.96 By working with over 26 mobile operators around the world, Cubic can remove the need for a
manufacturer to agree separate deals with multiple operators who might each have different
protocols. Car manufacturer Audi made a deal with Cubic in 2016 to support its connected car roll
out.97
There are a number of other difficulties that may also need to be overcome in this example, such as
billing complexity – who will pay for the connectivity at the point of use?
To the extent that connectivity is “necessary” for the car or is included for the manufactures benefit (for
example remote diagnostics), then the manufacturer might have to pay for some of the costs of the
connectivity. For those services that are purely add-ons (e.g. some infotainment services like using data
for in car streaming of entertainment), then the consumer might have to pay extra on a regular basis
for the data it uses.
To the extent that 5G creates services and products with embedded

connectivity that will be sold trans-nationally, there will be similar
opportunities for such aggregators in other industries too. This
might reduce transaction costs and also enhance the bargaining
power of users relative to network operators by combining the
demand of several users. These trans-national aggregators are
probably unlikely to be network operators given that none currently
have truly global reach, and could be considered somewhat similar
to a pan-national MVNO.
Given the potentially diverse needs of different sectors and users,
this is just one possible role for intermediaries downstream of
network operators. For example, 5G may create new business
opportunities for intermediaries to bundle and repackage
connectivity. There may be further opportunities for partnerships
95
Wired, “Ford links-up with Vodafone to bring 4G to European cars”, 27 February
2017. Available at: http://www.wired.co.uk/article/ford-in-car-wi-fi-modem-
vodafone-europe
96
See http://www.cubictelecom.com/
97
Cubic Telecom, “Audi and Cubic. Driving in-car connectivity”. Available at:
http://www.cubictelecom.com/Projects/Audi
63
between providers and users both to allow development of niche

services; this may help in understanding users’ particular
requirements and also overcoming the difficulty of creating
charging structures that work efficiently for different types of users.
If 5G opens up many niche applications, there may also be a role for
intermediaries with specialist knowledge of the specific connectivity
requirements of particular sectors. This could have important
consequences in terms of the kind of players regulators will have to
deal with in the future.
Network related
Network Typically when deploying a mobile network there may be a number
densification of alternative sites where operators could deploy their infrastructure
and still meet coverage and capacity requirements. However, for
the deployment of 5G networks, a greater number of small sites
within urban areas are likely to be needed (especially where
mmWave bands are used to support greater speed and capacity,
but for which the propagation characteristics of the spectrum mean
only relatively small areas can be covered per cell site). There will
also likely be a need for small-cells to be deployed within buildings
to serve indoor public spaces or privately controlled quasi-public
spaces (for example within shopping centres or at large stadia).
These sites will likely be privately owned and might be scarce,
which raises interesting questions around who will install the
infrastructure and how access will be granted (by which we mean
access to the physical site, and potentially also wholesale access to
the cell infrastructure).
For example, individual MNOs might deploy their own
infrastructure (i.e. small cells) in private sites. However, the
likelihood of multiple operators deploying in a single site might be
challenging given the need to reach agreements on access to
private sites (e.g. agreements with councils) and the practical
difficulties of providing power and backhaul to each site. Therefore,
there may only be scope for a single operator to gain access (or the
site owner may only offer exclusive access). The operator may (or
may not) then offer wholesale access for other operators. This is
illustrated in ‘New model 1’ in Figure 12 below.
64
Figure 12: The value chain considering cell densification and access to small
sites
At present (macro sites): With 5G and sites for small-cells With 5G and sites for small-cells
New model 1: New model 2:
Small-site Small-site
(privately owned) (privately owned)
Alternative
Site
Site
Physical access controlled Physical access controlled
Exclusive access
Exclusive access to site
to site
Intermediary
MNO (neutral host)
MNO MNO
Wholesale Wholesale
capacity capacity
M(V)NO MNO MNO

MVNO
MVNO
However, there may also be opportunities for new players upstream

of traditional mobile networks. Such intermediaries may wish to
acquire sites in dense urban areas and indoor public spaces, deploy
their own 5G infrastructure and offer wholesale services providing
patches of connectivity to 5G operators This is illustrated in ‘New
model 2’ in Figure 12 above.
We are already seeing businesses forming with such strategies
aimed at creating wholesale 5G capacity to sell to MNOs in order to
aid cell densification. For example, an operator called Airspan
recently won spectrum in the Irish 3.6 GHz award, apparently with a
view to providing small-cells and offering wholesale access.
In the case of gaining access to private sites for deployment of cells,
the role of the site owner in the value chain will be increasingly
important. There is a potential for the site owner to leverage its
bottleneck control in some cases where it might not be feasible or
cost effective for all public networks to place networks within that
space. To the extent that such issues might be a barrier to 5G
deployment we discuss them in more detail with the help of some
illustrative examples in Section 6 and consider the regulatory
implications in Section 7.
Convergence of Typically, MNOs have served mobile broadband customers over

PPDR98-MNO-FWA commercial radio access networks, with other specialised providers
provision
servicing the Fixed Wireless Access (FWA) Market to serve rural
areas with fixed broadband. However, given the enhanced
capabilities of 5G offering a viable substitute to fixed networks, we
98
At present a separate network may be needed given for higher quality in terms of
redundancy, uninterruptable power supply, outage periods and coverage,
65
may see MNOs extending into the FWA market, especially where
this provides the opportunity for a new revenue source that could
help support the 5G business case.
This could be an important factor in countries where fixed
broadband is particularly expensive or difficult to deploy due to
rural populations or challenging geographies (e.g. Ireland,
Switzerland). As mobile broadband speeds increase relative to fixed
connections, consumers may increasingly substitute fixed
broadband services with mobile broadband services; this may
fundamentally change the nature of competition, eventually
necessitating a re-think of regulation of fixed services, as the local
loop could for the first time become contestable by wireless
services.
Furthermore, for Public Protection and Disaster Recovery (PPDR)
services currently offered over an independent network
infrastructure may be offered over the same network infrastructure
as commercial networks using SDN and NFV to effectively allow for
the creation of separate networks that share one physical
infrastructure, with each network designed to meet its own specific
requirements. This would allow for significant cost savings to the
MNOs through reduced infrastructure duplication. Therefore, 5G
may affect the strategy for future PPDR services.
As illustrated in Figure 13 below, we may see some consolidation
across these sectors.
Figure 13: Convergence of PPDR-MNO-FWA provision
At present:
Spectrum
Synergy
PPDR MNO FWA
Emergency
services
With 5G:
Spectrum
Consolidation Consolidation
Synergy
PPDR MNO FWA
Emergency
services
66
Infrastructure For some use cases, it may also be attractive for 5G providers to
sharing enter into infrastructure sharing agreements to lower the costs of
deployment and avoid unnecessary and costly infrastructure
duplication. This might be made easier in a 5G world if SDN and NFV
could be used to “effectively create separate networks that are housed
within one physical infrastructure in a way that is tantamount to them
being situated on separate physical infrastructures. In this way, each
‘physical network’ will be able to host multiple service providers who
provide specialist niche services over that network”99
Cost and benefit There may be scope for agreements between network providers
sharing... and verticals requiring connectivity to share costs and risks. More
broadly, if there are incentive misalignments between verticals and
network operators, these might be addressed through co-
investment or Joint Ventures. These partnerships may be needed to
support network development or service development
…to support the For example, where network investment and 5G deployment would
network investment be very costly or unattractive, but verticals foresee significant
advantage associated with 5G connectivity (i.e. there is a large
private value for connectivity), then they may be willing to support
the MNO investment under co-investment models. However, given
that the requirements for connectivity might only be for particular
areas or be quite niche, it may be unlikely that such arrangements (if
made at all) would contribute significantly to the widespread roll-
out of 5G networks, for example at a national level.
…to support service Where MNOs are looking to provide tailored services to a particular
development use-case of a vertical but consider demand is uncertain, making
long term agreements could underpin the business case and limit
risks (with the connectivity providers sharing benefits in terms of
incremental revenues or reduced costs as a reward for a long-term
commitment). However, the extent to which this will occur depends
on the extent to which the vertical will need connectivity directly
from the MNO to meet its requirements or the extent to which it
could be served by intermediaries, as discussed above.
Will these changes be realised?

Although there is a wide range of possible changes facilitated by
the emergence of 5G, the extent to which these changes will be
99
Peter Alexiadis and Tony Shortall, “The Advent of 5G: Should Technological
Evolution Lead to Regulatory Revolution?” CPI Antitrust Chronicle November, Vol 3,
Autumn 2016. Available at:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2876484
67
realised in practice will depend on the widespread deployment and

take up of 5G.
Deployment of 5G networks itself will depend on a large number of
factors; not least the costs of investment, the expected demand,
incremental revenue raising opportunities and any potential
network cost savings (unit costs of network capacity might be lower
on 5G than 4G networks). In turn, take-up will depend on the extent
to which MNOs (or other connectivity providers) can successfully
identify and market services to customers who might find 5G useful
and the relative pricing of 5G and alternative technologies.
In the following section, we discuss why we believe that in the short
term the emergence of 5G will result in a continuation of the status
quo with the main driver and use case being provision of eMBB. We
then discuss the extent to which, over time, the 5G business case
may be supported by other use cases which will take advantage of
5G capabilities and support some of the new business models
described above. As part of this assessment we consider what
hurdles might need to be overcome to support the emergence of
5G.
68
Drivers and obstacles to 5G
6 Drivers and obstacles to 5G
6.1 Enhanced mobile broadband as the initial

driver
One of the key expected benefits of 5G that will drive early
deployment is its ability to provide eMBB to support the expected
increase in mobile data.
Continued data Mobile data traffic continues to increase across Europe despite
growth is expected significant variation in data consumption patterns across networks,
markets and subscriber segments. Ericsson reported that total
mobile data traffic has grown by 70% year-on-year between 2012
and 2017 globally and is expected to grow with a compound annual
growth rate (CAGR) of around 42 percent between 2017 and
2023100. This growth is not only driven by the rising number of
smartphone subscriptions, but also by changes in data
consumption patterns. Average data volume per subscription has
increased substantially over the past years, fuelled primarily by
more viewing of video content. Video streaming is expected to
grow to about 75% of all mobile traffic globally by 2023 at the
expense of the share of data dedicated to web browsing.101
Of course, exponential growth cannot continue indefinitely if the
main driver continues to be from video on mobile devices. Growth
forecasts must be treated with some scepticism, as ultimately
growth from smartphones will have to subside. 102 However, in the
long-run there may also be new sources of growth, such as
machine-to-machine communications which will form an increasing
share of total data traffic.
100
Ericsson, “Ericsson Mobility Report on the pulse of the networked society”,
November 2017.
101
Ericsson, “Ericsson Mobility Report on the pulse of the networked society”,
November 2017.
102
For example, much growth to date has been driven by video which can already
be provided in HD on many consumer devices. The user experience on these
devices are unlikely to benefit from further improvements, say to 4K HDR video.
Therefore, although data requirement may continue to rise, sustaining these
compound growth rates is not credible as it implies unreasonable levels of data
consumption.
69
5G can provide Given its capabilities that allow for improvements in spectral
network capacity efficiency, higher capacity and improved performance, migration to
more cheaply and
5G will be important in meeting further traffic growth cost
maintain or improve
service quality in the effectively and ensuring that service quality does not degrade
face of traffic growth leaving network operators at a disadvantage relative to their
competitors.103
Therefore, in the short to medium term, telecom operators
deploying 5G will likely focus on eMBB. Interviews conducted as
part of this study suggest that eMBB is the priority in these early
stages of 5G. Telecom operators responding in the Ericsson ‘5G
Readiness Survey’104 also suggested “taking market share from
competitors based on new features/performance” and “migrating
current 4G subscribers to 5G with enhanced revenues” were the
preferred means of monetising 5G connectivity. Furthermore, MBB
is an existing mass-market service and is not subject to significant
uncertainties around the size of markets that new services face.
eMBB and network The launch of eMBB over 5G can be thought of as the ‘evolutionary’
cost saving are the aspect of 5G, with 5G providing incremental improvements over 4G
initial drivers of 5G
in meeting data traffic growth. For example, as noted in the EC
deployment
5GAP Staff Working Paper,105 5G will first support and then replace
4G and therefore has an important role in bringing incremental
improvements to existing services: “5G is not conceived as a
technology replacing 4G, but rather enhancing it and complementing it
with new services capabilities. At this time, it is considered that the
usage of 4G will continue for many years, before eventually 5G takes
over completely…” and “…this evolutionary perspective is a key aspect
to support gradual investments in 5G, offering an introduction
perspective designed to gradually complement the 4G offer”.
6.1.1 Potential obstacles to deploying 5G networks for

eMBB
Small sites become Whilst eMBB may be the main initial driver for 5G, deployment of
important for infrastructure to support eMBB raises some logistical challenges and
capacity and to
potential competition concerns arising from control of access to
provide indoor
coverage sites that regulators should be aware of. In particular, challenges
associated with the need for a larger number of small sites and
103
This is similar to the incentives observed for 4G deployment, where lowering the
unit costs of network capacity as data usage grew was a key driver.
104
the road to 5G”.
105
Commission Staff Working Document, 5G Global Developments Accompanying
The Document, Communication From The Commission To The European
Parliament, The Council, The European Economic And Social Committee And The
Committee Of The Regions, 5G For Europe: An Action Plan.
70
deployment of 5G indoors will become more prevalent with 5G

than with previous technologies, not least due to the short-range
propagation characteristics of the very high frequency spectrum
(e.g. the mmWave bands). Ensuring indoor coverage will likely
become an increasingly important issue over time, with problems
likely to get worse rather than better due to tougher energy
efficiency standards for buildings. As we discuss in the boxes below,
technological changes may affect who has the upper hand in the
value chain, with more control being put in the hands of the site
owners. For example, there is potential for bottleneck control in
some cases where it might not be feasible or cost effective for all
public networks to place separate network infrastructure within that
space.
There are two scenarios that might become commonplace in the
roll out of 5G networks to support eMBB services:
• dense network deployment for public networks in urban
environments where very high frequency mmWave bands
will be needed to support capacity requirements; and
• provision of eMBB services within privately-owned but
‘quasi-public’ spaces (such as shopping malls, or large
stadia).
We discuss these situations in the boxes below and discuss the
regulatory implications in more detail in Section 7.
71
Example: 5G deployment for eMBB in urban environments.

Given the typically short range propagation of 5G small cells, multiple 5G small cells may need to be
deployed to give contiguous coverage and configured to allow hand over between cells in order to
provide consistent performance for users on the move. As demand and data consumption increases
over time more small cells will need to be added.
MNOs will want to ensure that they can deploy a sufficient number of small cells in a suitable topology
to ensure contiguous coverage within particular data hot spots such as city centres. However, this will
typically require locating equipment on pre-existing public or private infrastructure. To negotiate
access to each site would be potentially burdensome.
The proposals published in the 2016 Electronic Communications Code, imply ‘competent authorities’
should do as much as possible to help (or at least not unduly restrict deployment).106 It may be that,
recognising the needs of MNOs to get access to a large number of sites, local councils could offer
access to public infrastructure (for example street lamps). They may do so on a tender basis, seeking to
use their position as site owners to extract rent from telecoms companies looking to deploy. Given the
costs of deployment and the need for power and backhaul to each cell, it might be cost efficient for
only one supplier to deploy the network and then allow access to others, and this will need to be
accounted for in the tender process. However, public authorities may often have strong incentives to
raise revenue from commercial activities to supplement public funding, so a council might seek to
maximise its returns, for example by granting one MNO exclusive access for a fee reflecting the
commercial benefits to the MNO in terms of reduced cost and enhanced service quality relative to its
rivals. Therefore, the site owner might use its power to distort competition in mobile sectors and look
to extract a financial benefit from this.
One might hope that the site owner could specify the need for wholesale access to be granted to other
MNOs as part of the tender. This would open up the possibility for the provider to be an incumbent
telecoms operator or an independent ‘neutral host’. However, if the incentives of the site owner are to
grant exclusive access, there may be a need for regulatory intervention to ensure that other 5G
providers can gain access to the site to provide services too. A general obligation to provide cost-
oriented wholesale access might be difficult given that the costs may vary significantly from one site to
the next. It might be possible to introduce an obligation to provide access on reasonable terms in a
similar way to that previously introduced for site/mast sharing.
106
In the current proposals for the ECC (which may, of course, be revised) Article 56
of the ECC was added to cover: “Deployment and operation of small-area wireless
access points” and states: “Competent authorities shall allow the deployment,
connection and operation of unobtrusive small-area wireless access points under the
general authorisation regime and shall not unduly restrict that deployment, connection
or operation through individual town planning permits or in any other way…” See
Proposal for a DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL
establishing the European Electronic Communications Code (Recast), Brussels,
12.10.2016 available at: http://eur-
lex.europa.eu/resource.html?uri=cellar:c5ee8d55-7a56-11e6-b076-
01aa75ed71a1.0001.02/DOC_3&format=PDF
72
Example: Provision of public 5G networks for eMBB within private spaces.

Take a simple example of a shopping mall or an events venue such as a stadium where there might be
significant demand for access to mobile broadband by a large number of users simultaneously
requiring high capacity networks to support the demand.
This ‘in-building’ coverage requirement may not be met by the initial 5G network roll-outs (given the
propagation characteristics of spectrum – particularly very high frequency needed for capacity – limit
the extent to which signals will penetrate building to provide indoor coverage). Therefore, to provide
sufficient coverage, there will be a requirement for specific infrastructure deployed in the building
(such as small-cells).
There are a number of possible ways this could be provided. For example: Telecoms providers may like
to offer “boosted” coverage on the site, as part of their value proposition to consumers; a site owner
may want to procure a network solution from a single party; or the site owner may wish to self provide.
The self-provision would require deploying its own infrastructure and making use of shared spectrum
(or even trying to get ownership of specific spectrum allocations) and then provide wholesale access to
other MNOs, potentially exploiting its bottleneck control.
In the case of multiple telecoms operators wanting to deploy their own infrastructure/service, this
would potentially involve the deployment of a very large number of different small cell sites by each
MNO that wants to offer this service. This would not only be costly, but involve inefficient duplication
of infrastructure, notwithstanding that it may also be very difficult for all parties to get permission to
install such cells and meet other logistical challenges such as provision of power and connecting to
backhaul.
Recognising this, and realising that it could exploit its bottleneck control and extract potentially
significant rents from service providers, the site owner could tender for one party to install the
necessary infrastructure and provide the services on an exclusive basis. This might certainly be an
attractive proposition for some MNOs in the early days of 5G where there may be competition on
coverage (for example, as we saw in the early stages of 4G deployment).
However, there may be additional complexity in this case where interoperability is important (e.g. all
customers attending their event/site should have improved access regardless of provider). Similar
issues to those described in the box above will apply:
• How might the site owner exploit its position?
• How can it ensure wholesale access will be provided without distorting incentives?
• Is there a role for regulators in enforcing/controlling the need for wholesale access?
• How should access be mandated or controlled?
MIMO can be used to In an interview with one telecoms operator, it was recognised that
upgrade existing network densification will be required to provide high speed, high
sites and put off
acquiring sites for
capacity services in future, but mentioned that in the first instance it
small cells will want to make as much use as possible of existing sites and
assets. Therefore, use of MIMO technologies at existing sites may be
preferable in the short term. Small cells may be an option for further
network roll-out but this might only come later given the issues
associated with gaining access to sites and having to deploy fibre
backhaul and provide power. Given these hurdles, it is important
that issues such as planning are dealt with effectively so as not to
hinder the roll-out of networks necessary to ensure suitable
coverage in dense areas and indoors.
73
We note that for such sites access to fibre backhaul may also be
required107 (which depends on the availability of fibre networks) and
therefore access to fixed infrastructure (such as ducts and poles)
may assist small-cell deployments. Where such arrangements are
not already in place, NRAs may need to consider the terms on which
access should be made available.
Fixed networks may There may be other roll-out options for operators with both fixed
have additional and mobile assets. For example, in Spain, Telefónica is considering
options for small cell
rolling out a new generation of routers for its Movistar-branded
deployment
residential fibre-optic service that will incorporate a mobile antenna
to support 5G roll-out. Although the routers are currently at the
prototype stage, if this could be achieved this might provide an
alternative option for some telecom operators looking to reduce the
cost of installation of small cells on private sites.108 Note that this
might suggest there could be some deployment advantages to
those telecoms operators with both a fixed and mobile service
offering who can take advantage of supply-side complementarities.
6.1.2 Will eMBB lead to sufficient incremental revenue to

support the business case?
There is limited Although eMBB may be the initial driver of 5G deployment, it is
additional revenue unclear if consumers will be willing to pay significant more for the
from eMBB
enhanced broadband, especially if there are initially limited
applications/services that can truly take advantage of it.
Initially, 5G is likely to lead to higher data rates and better consumer
experiences for existing services. However, history suggests that
there will not be a significant willingness to pay a premium for
higher data rate services over mobile broadband. For example, not
even 4G deployments have been able to turn around the declining
ARPUs for mobile operators in Europe109.
107
However, non-fibre alternative technologies may also be implemented to
address the challenges for the backhaul of 5G networks. For example, in February
2018, Vodafone and Huawei have announced the completion of lab tests which
indicate that traditional IP microwave links will be considered a viable technology
for 5G backhaul: http://www.vodafone.com/content/index/what/technology-
blog/vodafone-and-huawei-test-applicability-of-ip-microwave-backhaul-for-
5g.html
108
Telecom Paper, “Telefonica testing new home fibre router with 5G antenna-
report”, 23 January 2018. Available at:
https://www.telecompaper.com/news/telefonica-testing-new-home-fibre-router-
with-5g-antenna-report--1228679
109
Analysys Mason, Report for Qualcomm, “Study on regulatory options to
promote investment in 5G and IoT infrastructure in Europe”, 15 December 2016.
74
Consumers get more Although consumers are getting more for their money when it
over time for a comes to mobile broadband packages (i.e. a greater data allowance
broadly similar cost
within the bundle), the overall spend remains relatively unchanged,
for mobile data
subscriptions in real suggesting that willingness to pay for the entire service is a limiting
terms factor.
The essential limitation on incremental revenues from eMBB is what
people are willing to pay for their mobile service as a whole and
history suggests that the period over which a premium can be
charged for new technology (e.g. premium over existing 4G
services110) may be relatively short-lived before it is competed away
and prices drop back to the previous level.111 With limited sustained
revenue impact, eMBB services provided over 5G will have to pay
for itself through other means.
Indeed, a key driver of 5G deployment for eMBB will be cost savings,
primarily driven by a reduction in unit network costs of provision of
data services as a result of being able to support increased capacity
and spectral efficiency.
As we have seen with 4G, the initial driver was for improved mobile
broadband (primarily the support of video streaming over the
mobile network in the case of 4G). Whilst there was a short period
where a premium could be achieved, the main driver was the
reduction in the unit cost of capacity to support improved mobile
services. This seems to have been the trend in recent years, with the
benefits of improved mobile broadband networks arising from cost
reductions and efficiencies rather than from improved revenues. For
example, although customers are spending a similar amount for a
mobile package, they typically get more for it i.e. the unit cost is
falling. This is demonstrated by recent findings by the Office for
National Statistics in the UK that although prices for telecoms
services in the UK were relatively flat between 2010 and 2015, the
way in which those prices are measured does not account for
110
Although there are some examples of premia being charged for 4G-capable
tarffis when 4G was first being deployed, it is unclear whether it would even be
possible to change a premium for specifically 5G eMBB under the Net Neutral
Regulation (EC2015/2120).
111
For example, a 2014 4G LTE Pricing Strategies report reviewing pricing in EU,
non-EU, Asia Pacific & Middle East regions shows that, “Although 4G services were
launched as a price premium by most MNOs, by 2014 the majority of 4G LTE plans are
now integrated within a broader mobile data tariff structure with pricing based mainly
on monthly data allowance, but with some MNOs offering higher access speeds if the
user opts for a larger data allowance. Rather than use 4G LTE services as a premium
service most MNOs are now offering 4G LTE for free to maintain an existing mobile data
price point. Price stability is now the objective rather than premium pricing”. See:
http://www.telecomspricing.com/news_detail.cfm?item=8594
75
improvements in the bundles of calls, texts and minutes meaning

that unit prices are falling.112
Incremental Therefore, although there will be incentives for network operators
revenues are likely to deploy 5G to reduce network costs in meeting data growth from
form new services
mobile broadband services, it is not obvious that traditional
and customers over
the longer run, not customer relationships can deliver significant and sustained
from eMBB incremental revenue growth as a result of eMBB alone. Rather,
incremental sources of revenue are likely to be from new customers;
in the longer run, new uses - other than mobile broadband services
- may be important to delivering sufficient return on investment in
from 5G investments.
6.2 Identifying new revenue sources from

additional capabilities of 5G
With little sustained incremental revenues from traditional
customers from 5G eMBB, MNOs (and other 5G providers) might
seek to find new incremental revenue sources by supporting
developments in other industries.
5G allows 5G has particular characteristics that could allow it to support
differentiated multiple different uses over a single network. For example,
services that might
SDN/NFV gives network operators agility to create new services
expand demand by
attracting new without the need for changing the underlying physical
customers with new infrastructure. This will provide a more cost-efficient approach to
uses adapting networks to meet evolving demands for connectivity and
also allow for a much greater degree of differentiation of services.
For example, a common infrastructure can use network slices to
support multiple services that each have a different priority, such as
bandwidth, reliability, terminal density or low power usage.
However, the extent to which 5G use cases beyond eMBB will be
realised within the timeframe of this study will depend on the
demand side, and the willingness to pay for 5G services, which will
in part be influenced by availability and price of alternative
solutions.
As shown earlier with our discussions in a number of sectors, it is
very difficult to say whether any one sector/vertical will or will not
need 5G going forward. Many use-cases may be possible with
alternative technologies or simply evolve towards reliance on 5G
when it becomes available.
112
For example, see Financial Times, “ONS’s crossed telecom wires raise questions
over inflation figures”, 18 Jan 2018. Available at:
https://www.ft.com/content/abc14c66-fb78-11e7-a492-2c9be7f3120a
76
New users arising Therefore, rather than thinking about a small number of specific use
from 5G are likely to cases (for which the likely scale of any particular one alone might be
be fragmented in
insufficient to drive 5G deployment) a better way of looking at this
niches across many
sectors would be to think of particular niches that are relevant for 5G (e.g.
need for dense sensor networks) with a large number of use-cases
and beneficiaries fragmented over a wide number of sectors.
For example, consider key 5G characteristics that certain niche use
cases will require (potentially in a large number of industries):
• Low latency will be crucial for real-time control and safety
critical applications (and in many of these cases mobile
edge computing will be important);
• Security/reliability/dedicated capacity may also be
important for some particular niches requiring service level
guarantees (SLGs) that may be met by “ultra-reliable” 5G
(although 5G may be one of multiple technologies sourced
for redundancy purposes)
• Connection density and network energy efficiency
(supporting low power devices) will be important for IoT
and mMTC across a wide range of sectors;
• Peak bandwidth requirements where existing wireless
networks may struggle (especially as competing demands
for bandwidth emerge).
In the table below, we give some possible examples for each of
these referring to those use cases touched upon in Section 4.
77
Table 9: Examples from case studies
Media and
Automo- Manu- Agricul- Energy and Other
enterta- Logistics Health-care
tive facturing ture utilities verticals
inment
Low Safety AR/VR AR/VR Drones Drones Wireless Real time

latency applications Telesurgery control
Off-site Critical M2M applications
media communicat
production ion/safety
Content Critical M2M Drones Drones Smart grids Wireless

SLGs / Safety distribution communicat Telesurgery Safety
reserved applications ion/safety applications
capacity
Public safety
Stadium/ Sensor Sensor Asset Dense

Device events networks/ networks/ tracking sensor
Density M2M M2M networks for
M2M or IoT
Sensor IoT tracking Sensor Smart Dense

Low power networks/ networks wearables sensor
use M2M networks for
M2M or IoT
Peak Live video Immersive AR/VR Wireless

bandwidth for ‘see media such telesurgery
through the as AR/VR
front’
vehicle 360 video
Incremental Together a potentially large number beneficiaries over a wide

revenues from 5G are number of sectors could be sufficient to support a significant
hard to forecast
market for 5G. However, the specific niches and fragmentation
across industries mean it will be difficult for any investor to estimate
the exact demand for 5G services. This uncertainty makes the
business case for 5G investments challenging. The difficulty lies in
being able to identify the demand and market services to these
users.
6.2.1 Potential obstacles to generating incremental

revenues from new sources
A new service might In some cases, the demand for 5G services will be influenced purely
be dependent on 5G by technical requirements that can only be met with 5G (i.e. a
revolutionary new service that cannot be done with other wireless
technologies). Whilst, there may be some use cases that genuinely
require 5G that could not be supported with other technologies, we
have not come across any compelling evidence to convince us that
such ‘revolutionary’ services relying solely on 5G will be available
within the shorter timeframe of this study and so affect near term
5G roll-out decisions. If a service is to rely solely on 5G then the
emergence of such services will likely be delayed until 5G
deployments are much more developed and widespread.
78
It is more likely that There is clearly a high level of risk for developers working on
services, applications developing applications or services that might benefit from 5G, as it
and products will be
cannot be assumed that 5G will be ubiquitously available for some
built for a mixed
connectivity time. Rather for some considerable time there will be a mixed
environment connectivity environment (with 5G running in parallel with 4G and
RLAN often providing indoor connectivity). Therefore, commercial
services and products using mobile connectivity will typically need
to be developed with fall-back options rather relying entirely on 5G.
If fall back Therefore, given that we expect most services that will emerge in
connectivity options the short to medium term will be designed to operate over a range
are needed, they also
of technologies (of which 5G may be one), 5G will need to be
act as alternatives
that limit willingness attractive relative to other technological solutions. Whilst 5G has
to pay for 5G potential to provide enhancements, the amount users will be
prepared to pay additionally for 5G functionality will depend on the
price and availability of alternative technologies. For many, the
decision to use 5G will depend on the relative pricing and the
incremental benefits relative to existing solutions that could meet
their requirements i.e. the use of 5G will be a cost efficiency
decision (so the charging model will have an important role in
expanding the market).
Take up of 5G based on cost efficiency considerations

One example of the need for 5G to compete with existing
technologies, and an issue likely to be common across a large
number of sectors, is the expected increase in IoT and mMTC
applications. The large number of potential use-cases for IoT and
mMTC suggests that there will be a significant new market to be
served with great potential for MNOs to generate incremental
revenue. Given the expected size of the market for these
applications in future, this could provide a significant further means
of monetizing 5G deployment. For example, Alexiadis and Shortall
(2016) suggested that: “No longer will voice communications become
the primary revenue source for operators, nor will they account for the
bulk of communications. Instead, machines communicating with
machines (e.g. self-driving cars) will provide the momentum for the
business case in the sector”.113
However, MNOs (or other providers) wishing to market their 5G
offering to support increasing demand for embedded connectivity
in devices and sensor networks will have to compete with
alternative technologies. Many of these use cases may be supported
113
Peter Alexiadis and Tony Shortall, “The Advent of 5G: Should Technological
Evolution Lead to Regulatory Revolution?” CPI Antitrust Chronicle November, Vol 3,
Autumn 2016. Available at:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2876484
79
by other technologies that are fairly mature. Therefore, the extent to

which these use cases take up 5G will depend on charging
structure for 5G services.
Some charging In the box below, we consider some of the specific issues with a
structures might be simple example linked to the case of deploying a dense sensor
inefficient as they
network. Whilst we consider this in the context of agriculture, one
create poor
incentives for can imagine that many of the key issues will translate across to a
potential users large number of IoT or M2M use-cases reliant on a large number of
sensors communicating with each other.
Example: Crop monitoring service for agriculture with dense sensor networks
Take an example of a provider (referred to as the ‘service provider’) offering a service to the agricultural
sector which allows them to gather and interpret data from a large number of sensors around the farm
which allow them to make more efficient decisions around where to use fertiliser, irrigate and so on.
This service provider might offer an integrated solution, which includes the sensors and analytics
software together with a ‘connected’ solution to allow everything to communicate with each other. For
the service provider (and the farmer) the primarily value of the service might lie in the analytics, but for
this to work well the sensor deployment needs to be integrated into the service offering.
Given that the bandwidth demands of each sensor are quite low, such a service could be provided
using existing technologies. For example, they may be able to deploy the sensor network with a “hub
and spoke topology”, where a number of ‘slave’ devices connect to a ‘master’ using NB-IoT
communication networks , with only that master device being connected to the mobile network using
a normal SIM to 4G or 3G network, which then sends the data to servers to be analysed and interpreted.
However, existing solutions can, in some cases, have some disadvantages that could be addressed by
5G. For example, using a spoke-hub or master-slave configuration can take time to market and require
professional design/configuration of the topology. Furthermore, as the number of devices increases
there may be some problems regarding devices density, since increasing the number of devices hinder
the topology design, organisation of communication of all devices, etc.
Developments in 4G and ultimately 5G may be able to help with this. e.g. in the case of agriculture: if
sensors would be independently connected (with low power consumption) to operator’s network, it
would be as easy as sell them to the field owner who can him/herself deploy the nodes where they
want, with no configuration at all (plug&play)114 with 5G also having advantages in terms of being able
to handle an increased density of devices and support low power devices.
The question is whether a cost effective solution can be sought for these “independently connected”
sensors, especially where there is a very large number? Therefore, the pricing and charging structure of
the 5G ‘solution’ will be an important factor in generating demand for using 5G networks for such
purposes, especially where the incremental benefit of 5G may be limited (i.e. not essential for the
provision of the service).
The traditional model of requiring a separate subscription per device and charging for data use on a
per device basis would be unlikely to be attractive/cost effective and therefore without changes to the
charging structure the market expanding possibilities of 5G may be limited. For example, if there are a
considerable number of devices, but each uses relatively little data there may be little willingness-to-
pay a relatively high marginal price to have a ‘mobile subscription’ for each device e.g. a farmer
installing hundreds/thousands or sensors could not be willing to pay a per-device cost unless the
marginal price is very low (e.g. some commercially available solutions offered by MNOs such as small
114
Based on our discussion with Codesian: https://codesian.com/
80
tracking devices are listed as costing £3 or £4 per month which would rapidly add up if large numbers
of devices were requires115 - albeit these are currently aimed at the consumer market).
However, if the marginal price per device can be low, then there is a potentially large market available
to the telecoms operator i.e. even if willingness to pay is only marginal, this may be summed over many
uses in many industries to create a substantial market for embedded connectivity.
Supposing that the MNO can establish a charging structure that would allow the agriculture service
provider to offer a service based on independently connected devices on a cost-effective basis (which
will make its proposition much more flexible and reduce deployment topology complexity etc.), how
will the service provider secure connectivity?
• will it do so on a per-site basis (i.e. having to negotiate a new deal for each and every site
depending on location and scale)?
• will it have to commit to a single MNO provider?
• will choice of provider depend on coverage (regionally and internationally)?
As discussed in section 5, cases like this raise the possibilities for new entrants or further changes to
business models. For example, given some of the issues involved with bi-lateral or multi-site
negotiations, there is potential scope for intermediaries to come into the market to facilitate deals with
telecoms companies across regions and boarders and provide a single solution to the end-user (i.e. the
service provider), operating a bit like an MVNO. Alternatively, if the potential market is large enough
the telecoms operator (which might even be a virtual operator using network slicing) may seek to
vertically integrate with the service provider, or there may be a need for specialist intermediaries who
can help better understand the market and provide tailored services.
Similar issues around pricing may also be relevant for other cases.
For example, the emergence of such use cases relying on low
latency and high bandwidth will depend to some degree on the
balance between of the cost of local computing against the costs of
connectivity. In some cases, it may be more cost efficient to have
more complex devices that do the necessary computation
themselves on the device. For example, control applications relying
on very low-latency could presumably use largely local compute
resources rather than relying on 5G. However, depending on the
pricing, devices might be ‘dumb’ and make use of 5G connectivity
to send the data to a centralised resource for computation and then
receive the outputs. So again, the charging model adopted by
telecoms operators will be key to the incentives of the user. For
example, charges do not necessarily have to be priced on a per bit
or capacity basis and instead, charges could be tailored, for example
for services that may emphasise quality of services metrics charges
could be differentiated by latency targets.
Self-provision of 5G connectivity
We have mainly focussed on the business case from the perspective
of an MNO on the basis that they will be looking to support new use
cases, seeking incremental revenues to support the 5G business
115
V by Vodafone. For examples, see: http://www.vodafone.co.uk/v-by-vodafone/
81
case. However, it may also be that for some particular use cases,
private networks would be a preferred solution.
Small cells can For example, consider a case where there are requirements for
support private 5G ‘connectivity’ within a single, well-defined environment to support
networks
IoT or mMTC (for example, within a factory). This ‘in building’
coverage requirement may not be met by the initial 5G network
roll-outs (as the propagation characteristics of spectrum –
particularly very high frequency needed for capacity – limit the
extent to which signals will penetrate building to provide indoor
coverage). Therefore, to ensure sufficient coverage, there will be a
requirement for specific infrastructure deployed in the building
(such as small cells).
This might be provided in a number of ways:
• MNOs offering to provide specific infrastructure and a
tailored 5G solution, seeing this as an opportunity to
diversify their standard service and expand demand by
providing a solution that the enterprise could ‘buy’; or
• depending on the specific requirements the enterprise may
instead consider deploying its own private network
solution.
In this case, the enterprise faces a decision of whether to “build or
buy” 5G connectivity.
82
Example: private network to support IoT M2M on site (e.g. a factory)

A factory owner may be able to (or currently does) provide an in-house private network using shared
spectrum and existing technologies (such as Wi-Fi solutions), to allow devices and sensors to
communicate within the factory. This may be particularly relevant as Wi-Fi networks continue to evolve
and become more capable. We understand that the IEEE is continuing to work with the 3GPP on ways
to meet IMT-2020 requirements and further Wi-Fi developments can be expected.
However, if the number of connected devices increases and the demands of those devices increase
(with respect to data throughput or low latency requirements for example), to a point that Wi-Fi based
solutions will have limitations in terms of ability to handle a large number of devices, support high
throughput, and provide sufficient coverage in a cost efficient way, the factory owner may seek to gain
connectivity through use of other wireless technologies that can handle their demands, such as 5G.
One option would be to source connectivity from MNOs. For example, the enterprise would be able to
procure connectivity from an MNO (or other ‘non-traditional’ connectivity provider with access to 5G
spectrum). The site owner could tender for one party to install the necessary infrastructure and provide
the services on an exclusive basis (similar to that discussed in the stadia/shopping mall case).
However, in some cases there may be reasons why an enterprise may prefer to ‘build’ rather than ‘buy’.
For example, depending on the extent to which there are connectivity requirements for other parts of
the factory (such as machine-to-machine communication, remote monitoring devices, or services
relying on mission-critical communications, for example), the manufacturer may demand some control
over the network rather than relying solely on an external provider or MNO. We understand that there
will be some clear cases where enterprises will have a strict preference for building their own private
5G network so as to remain in full control of the network and may therefore seek to self-supply by
installing specific infrastructure and acquire access to spectrum.116
When deploying private networks verticals may have some

difficulties, given they are not communications experts,117 and given
the need to access suitable spectrum.118 On the other hand, some
verticals may have already acquired good knowledge and
experience by developing and operating their own networks using
existing technologies.
Private 5G networks We consider that the deployment of private networks should not be
should not crowd discouraged, as there may be some benefits associated with such
incentives to roll out
public 5G
small-scale deployments. For example, in cases such as those
described above, the deployment of a private network would not
necessarily undermine the MNO business case for deployment of 5G
public networks, given that supply would in any case require
specific network investment beyond that required for the public
network. In other words, the private network would not be
crowding out investment in the public network.
116
This could be access to shared or secondary use spectrum. We discuss spectrum
issues in more detail in Section 7.
117
In an interview with one telecoms operator, it pointed out that verticals are not
communications experts and may build a network to meet their needs now, but
end up in a technology ‘cul-de-sac’ down the line, as they do not have the expertise
to be able to adapt their network to meet their changing needs.
118
We discuss the spectrum issues in section 7
83
Regulatory options
7 Regulatory options
In many ways 5G is evolutionary, rather than revolutionary.
Nevertheless, it may lead to significant changes, including:
• increased use of small cells to achieve higher data rates;
• possible entry of non-traditional network operators
focussing on cell densification;
• differentiation of service characteristics through network
slicing to meet the diverse needs of various ‘vertical’
industries and other user groups;
• possibilities for creating private networks (for example, in
industrial spaces, especially indoors);
• new opportunities for intermediation of various forms to
integrate different networks and provide connectivity
tailored to particular users’ needs; and
• low-latency edge computing being provided by network
operators.
In parallel with these developments, 4G coverage will continue to
increase, with 4G adding various capabilities in terms of speed and
its ability to serve IoT and M2M applications needing low power
terminals. Therefore, mobile networks will develop as a patchwork,
with a mix of 4G (LTE-A) and 5G coverage, plus integration of other
network infrastructure, such as use of fixed network infrastructure
to deploy small cells and wholesale services from neutral hosts (who
might be other MNOs).
In this section, we discuss what issues these changes might create
for national regulators. We are not seeking to make specific
recommendations, but rather to identify where attention is likely to
be needed from regulators given the likely trends.
7.1 Challenges arising from small cells

Small cells will be Higher data rates necessarily require smaller cells. Future use of
needed mmWave bands for 5G will require very small cells, including within
buildings and amongst street furniture. Suitable sites for small cells
are not easy to find. Lack of alternatives may give site owners
significant bargaining power relative to network operators.
This trend is largely inevitable as a result of the physical
characteristics of radio spectrum and growing demand for
bandwidth. It raises a number of challenges for regulators (and
governments more generally):
• the possibility of site owners making exclusive or
restrictive arrangements with network operators;
84
Regulatory options
• planning and other practical constraints on the supply of

locations for small cells; and
• the role of the public sector itself as a site owner within
urban environments.
Exclusive site access arrangements

A site owner wishing to maximise revenue might in some cases seek
to make an exclusive arrangement with a single network operator
or to exclude one or more operators. Regulators need to consider
how they might identify any competitive concerns and respond
proportionately. Providing a prior indication of the likely response
to such situations may help to discourage any anticompetitive
behaviour in the first place.
Exclusive or We would emphasise that exclusive or discriminatory site access is
discriminatory not a general problem, but rather one that may arise in particular
access to small cells
cases. Site owners will typically have incentives to make sites
is not a general
problem … accessible to all networks in order to maximise their revenues.
Nevertheless, in certain cases, it is possible that a site owner might
obtain greater revenue by limiting the number of network
operators at its site rather than accepting all interested parties. In
the most extreme case, this might be an exclusive access
arrangement, but there are other possibilities. For example, there
might be 4 competing MNOs, but the site owner might make only 3
(or fewer) access arrangements available; there would then be
competition amongst the MNOs to avoid being the one MNO
excluded from the site.
… but related to site Clearly if the site owner takes this approach it must earn sufficient
owners having additional revenue from each MNO it deals with to compensate for
sufficient power to
not dealing with all of MNOs. This could occur if either:
affect competition
• there is enough impact on downstream competition in
mobile markets due to only some MNOs having access to a
‘pivotal’ site; the site owner would then be extracting
revenue related to a restriction of downstream competition;
or
• the site owner is uncertain about the willingness of network
operators to pay for access to the site, and rather than either
negotiating individually with operators or posting a
standard price for access, seeks to create competition for
site access amongst network operators by excluding one or
more from the site.
Any impact on downstream competition is unlikely to come from
small sites, as only a small proportion of consumers will be affected
by the availability of services at that location. Network operators
may decide not to be present at that location if site access costs are
too high and their lack of access does not lead to a significant
competitive disadvantage in retail competition. Equally, small site
85
Regulatory options
owners are unlikely to adopt the strategy of committing to exclude

one or more networks to increase their bargaining power relative to
network operators, as the commitment is unlikely to be credible (i.e.
the site owner might let excluded networks in later to get additional
revenue later).
Concerns are largely Therefore, regulators need to be primarily concerned about large
limited to pivotal ‘pivotal’ sites or collections of sites under common control where
sites and large
access affects the perceived quality of mobile services for
collection of sites
under common consumers. For example, access to groups of small sites across a city
control (e.g. street lamps) or at major transport hubs may be important as a
competitive differentiator for MNOs; this might meet the conditions
necessary for the site owner to have an incentive to deal with a
subset of network operators (and possibly just one). Another
example might be a sufficiently large collection of sites, such as a
chain of major shopping malls.
Neutral hosts and If a competition problem arises how might this be remedied?
wholesale access Should obligations fall on network operators or site owners? There
obligations
are a number of possibilities:
• if one network operator is present, then it might be required
to offer access at a wholesale level to other network
operators over shared infrastructure;
• a variation would be a ‘neutral host’ who could install the
infrastructure and make wholesale access available on a
non-discriminatory basis to all network operators (though
the neutral host could be one particular MNO);
• an obligation could be imposed on the site owner to make
physical access available to all network operators on a non-
discriminatory basis.
Some NRA’s are already considering different approaches to this
issue. For example (as described in Section 2.4 above):
• Ofcom (UK) sought views on the potential for wholesale
access models.119
• RTR has indicated that it plans to allow infrastructure
sharing to facilitate densification of small cell deployment
for 5G roll-out.120
Difficulties in Even if a neutral host provides non-discriminatory access to all
constraining pivotal networks, this may be insufficient to constrain the exercise of
site owners who
might have market
market power by the site owner, if network operators have no
power alternative but to use that site. In some cases, especially with small
cells at millimetre wavelengths, network operators may not have
effective alternatives; this can be contrasted with the use of large
119
Ofcom, “Call for inputs on 5G spectrum access at 26 GHz and update on bands
above 30 GHz”, 28 July 2017.
120
RTR, “Consultation on the 3.4-3.8 GHz award procedure”, 28 July 2017
86
Regulatory options
cells for coverage at frequencies below 1 GHz, where there will be

greater choice in where to locate cell sites as signals will penetrate
buildings to some degree. This is a potentially difficult problem,
which regulators will need to keep under review as networks start
to rely on small cells at higher frequencies. There is no simple
regulatory solution, though competition law may constrain the
pricing of site access in such cases (either through ex post
complaints, or ex ante by providing incentives not to abuse market
power).
Obligations (for example, such as cost orientation) on pivotal site
owners are likely to be difficult to implement for a number of
reasons. First, the site owner might not be a telecoms operator and
so might not fall under ex ante regulatory obligations. Competition
law would nevertheless apply, but any intervention would need first
to demonstrate that the site owner had a dominant position in a
defined relevant market (i.e. the supply of physical access to
network operators at that location) and that this was being abused.
Second, it may be difficult to determine what a reasonable charge
for site access might be, as costs may be difficult to measure. Site
owners are likely to be building and maintaining infrastructure
whose primary purpose is not to host telecoms equipment.
Therefore, revenues from site rental are typically a contribution to
the site owner’s common and fixed costs. Site owners typically
negotiate with network operators and so charges are primarily a
reflection of network operators’ willingness to pay for access.
Third, the nature of small cells means that due to lack of space it
might not be physically possible to co-locate equipment for
multiple networks. Therefore, sharing of at least the RAN would be
needed in any case if multiple networks were to make use of the
site. This situation is likely to be common in practice, especially for
small cells within indoor, publicly-accessible spaces.
Site availability: planning and public sites

As discussed above, the move towards small cells (and towards
higher frequencies) may give site owners greater power within the
overall value chain. Measures aimed at ensuring that all networks
can gain access, possibly through neutral hosts, will minimise the
creation of any competitive asymmetries in downstream mobile
service markets. However, pivotal site owners may still remain in a
strong position. This might be detrimental if high site access costs
impede roll-out.
Increasing the supply Rather than trying to control any market power that pivotal site
of sites owners might have through competition law – which will be
complex given that each case will be different both in terms of the
extent of any market power and the specific costs to the site owner
– it may be easier to increase the supply of sites. There are a number
87
Regulatory options
of policy measures that could be taken, though these are questions

for broader government, rather than telecoms regulators:
• Planning rules need to reflect the changing nature of
infrastructure, with miniaturisation making it possible to
deploy small cells with limited visual impact;
• The public sector may control many suitable locations for
small cells within dense urban environments, such as street
furniture or roofs of social housing. Where public bodies
have objectives that include earning commercial revenue
and control pivotal cell site locations, there is a danger that
site access pricing could be excessive (just as if the site
owner were a private entity).
There is a real risk of Because of pressure on public finances, it is increasingly common
public bodies seeking for public bodies to have strong incentives to find commercial
exclusive access
revenue streams. It is a real possibility that this might create
deals to maximise
commercial revenue incentives to conclude exclusive site access arrangements.
By way of analogy, London Underground signed an exclusive
contract with Virgin Media to launch a Wi-Fi service at selected tube
stations on the London Underground network. After launch (June
2012) customers of all mobile networks were able to use Wi-Fi for
free during the Summer Olympics. After this period ended, only
Virgin Media customers were able to use the Wi-Fi as part of their
subscriptions. Customers of other networks using the Wi-Fi were
able to access travel information for free or purchase packages or
pay-as-you-go usage from Virgin Media. Since 2014 connecting to
the Wi-Fi has been free for EE, Vodafone, O2 and Three customers
after these operators signed wholesale agreements with Virgin
Media, with customers of other networks still able to purchase
access from Virgin Media.
7.2 Access to spectrum
Availability of spectrum for public networks

Spectrum at 3.6 GHz The recent “Second Opinion on 5G networks” from the RSPG121 has
and 26 GHz will be set out a roadmap for spectrum availability for 5G in which:
made available for
5G • The 3.4 – 3.8 GHz band is identified as the primary 5G band
for initial deployments;
• The 26 GHz band (24.25 GHz to 27.5 GHz) is identified as a
‘pioneer band’, primarily for individual licences rather than
“Strategic spectrum roadmap towards 5G for Europe, RSPG Second Opinion on

121
5G networks”, RSPG18-005, 30 January 2018.
88
Regulatory options
sharing, with a target for Member States to make a

sufficiently large portion of the band (e.g. 1 GHz) available
by 2020; and
• The 700 MHz band is designated to provide wide area
coverage.
At the time of writing, some auctions of 3.6 GHz band spectrum
have already been conducted; others are about to be run or are at
an advanced stage of planning. As a result of public sector spectrum
release programmes, much larger amounts of spectrum have been
made available in the 3.6 GHz band than in previous awards of
spectrum harmonised for mobile services. For example, 350 MHz of
spectrum was sold in the Irish award122 in 2017.
The 26 GHz band is at present commonly used for fixed links.
However, the frequencies identified by the RSPG (24.25 GHz to 27.5
GHz) are much wider than the typical fixed link allocations and in
many Member States parts of this band are not currently in use.
Therefore, notwithstanding any need to migrate fixed links in the
long run, the target set for release of this band in the RSPG’s second
opinion should be readily achievable. Even where migration of fixed
links might be needed, there are many possible bands already in
use for fixed links, so this should not be a major obstacle.
Spectrum for private 5G networks

Spectrum needs for Therefore, at present there are clear and credible plans for spectrum
private 5G networks to be made available in a timely manner for 5G on an individually
licensed basis. However, it is somewhat unclear what spectrum
might be appropriate to use for 5G within private spaces for private
networks, for example to support automation and control
applications within a large factory.
Coordination in One company we spoke to told us about an example with an
choice of bands for automated car parking solution, essentially using automated
private networks
trolleys requiring a private wireless network to control them.
However, they found that the way in which spectrum can be
accessed and used differs across member states making it difficult
to design a solution that is compatible across Europe. We were told
that Arcep decided that the 2.5 GHz band could be used to provide
a private network for such purpose. However, in other countries
there may not be such a framework for a private network and the
solution would require support from a network provider (e.g. under
the mobile operator licenced network). In Germany it was
122
ComReg, “Results of 3.6 GHz Band Spectrum Award”, 22 May 2017. Available at:
https://www.comreg.ie/publication/results-3-6-ghz-band-spectrum-award/
89
Regulatory options
determined the 3.6 GHz band should be used for such private
networks.
Clearly there may be a range of alternatives available for
applications requiring private networks. RLANs may be suitable for
some applications, especially given recent standards to support
faster speeds (WiGig) and power saving for IoT-type applications.
Narrowband telemetry and control applications might also be able
to use unlicensed spectrum (e.g. 433 MHz and 868 MHz). However,
all of these possibilities involve the use of shared spectrum on an
unprotected basis. We have heard concerns from potential users
about the reliability of using such sharing spectrum for mission-
critical applications given the possibility of interference considering
the large number of devices already in use on these unlicensed
bands. Therefore, there appears to be a requirement for private 5G
networks that are distinct from existing services and technologies
using shared spectrum.
Private networks as It would be helpful to clarify the licensing model for private 5G
secondary users networks and to identify which bands this might fall into. Shared
use does not appear to meet the requirements of potential users.
Equally, it might be a waste to reserve spectrum exclusively for
private users given that most uses of private networks would occur
indoors, using small cells at low power levels, thereby limiting
interference to other users outside. Therefore, there appears to be
scope to licence private networks as secondary users, sharing
spectrum on a non-interfering basis with public network operators
as primary users. For mmWave bands, the propagation
characteristics of the signals themselves greatly constrain the
potential for interference. Beamforming and massive MIMO can also
reduce interference by making signals more directional.
This secondary user model is different from unlicensed spectrum.
Any secondary user would require a licence to operate (as opposed
to unlicensed spectrum where interference is managed entirely
through restrictions of permitted equipment). The primary user
would be protected from interference from the secondary user. The
secondary user would need to operate within this constraint, but
would have some measure of protection, unlike shared spectrum.
This model would seem able to address the concerns that use of
shared spectrum is too risky for mission-critical applications. At the
same time, it avoids the need to commit large sections of bands for
providing highly localised, site-specific exclusive use licenses given
that the extent of demand for private 5G networks is uncertain. It
would highly inefficient if large ranges of frequencies were reserved
on a national basis to provide for a small number of private
networks covering a tiny geographical area.
Coordination of Clarity over the spectrum access model for private 5G networks is
approaches across needed at the EU level. However, this is not to say that detailed
Member States
harmonisation measures are needed. It would be sufficient simply
to clarify whether spectrum would be available for private networks
– we suggest licensed on a secondary basis – and, if so, which bands
90
Regulatory options
this would be available in. Such guidance would remove a potential

barrier to innovation in services due to cross-country differences.
This might be a matter on which BEREC could take a role in ensuring
reasonable consistency of licensing arrangements for private 5G
networks.
Taking a secondary licencing approach within spectrum bands used
for public 5G networks might also be preferable to making
spectrum available in distinct bands purely for 5G private networks
or to support innovative new use cases. Using the same spectrum
for public and private networks through sharing arrangements
removes a potential inhibitor to development if equipment
manufacturers cannot benefit from the scale economies they would
enjoy when supporting mass market services.
Accommodating non-traditional operators

New entry There is the possibility that 5G might encourage entry of new
possibilities network infrastructure providers. In particular, there may be
opportunity for specialist providers of cell densification offering
wholesale access to other network operators. Indeed, the 2017 Irish
3.6 GHz auction resulted in spectrum being awarded to Airspan,
who appear to have such a business model.123
We cannot judge at this point whether there will be sustainable
business models for cell densification by independent operators or
whether they might eventually consolidate into existing mobile
network operators. Nevertheless, it is possible that the need to
obtain suitable sites, conclude deals with site operators and then
interconnect and backhaul a large number of small cells within a
limited physical area could create opportunities for specialist
providers of wholesale services (acting as a neutral host).
Flexible spectrum Given this, spectrum award processes should be sufficiently flexible
packaging can meet to allow such entry by what we might call ‘non-traditional’
the needs of non-
operators. In particular, approaches in which spectrum is auctioned
traditional entrants
or awarded in pre-determined packages designed to meet the
needs of incumbent MNOs are inappropriate as they might inhibit
such entry. Recent spectrum auctions – regardless of the specific
auction format used – have tended to offer spectrum in small blocks
(typically 5 MHz), allowing competition over the amount of
spectrum acquired by bidders (subject to any spectrum caps). This
approach is sufficiently flexible to accommodate non-traditional
entrants, whose spectrum requirements may be difficult to
anticipate. With larger amounts of spectrum available (such as in
typically 3.6 GHz awards) it is possible that incumbent MNOs will
123
See Airspan Website: http://www.airspan.com
91
Regulatory options
not necessarily win all the available spectrum and entrants may be
successful, as in the Irish 3.6 GHz auction.
Licence conditions It is also important that licence conditions are not set in a manner
should not exclude that excludes non-traditional operators. For example, coverage
cell densification
obligations that might be easily met by an MNO with an existing
business cases
national network could be very onerous for an operator wanting to
offer wholesale services at certain key locations where it might not
be feasible for multiple networks to run dense deployments of small
cells. Put simply, cell densification by non-traditional operators
offering wholesale services may mean that coverage is ‘spotty’.
7.3 Coverage issues

General coverage is The 5G Action Plan124 sets targets for 5G availability in one major
important for city in each Member State by 2020. This will provide opportunities
innovation
to test 5G new applications and services. Whilst developers of new
incentives
services, applications and products using 5G will no doubt benefit
from this initiative in the short run, in the longer run their incentives
to innovate are much more related to how widespread 5G services
will be in general, which is not tied to any deployment plan for a
prioritised city.
Applications and No rational developer will want to take the risk of tying their
services will use a product or services entirely to 5G if coverage cannot be relied on.
mix of different
Rather, as we have heard in our interviews, products and services
forms of connectivity
will need to be able to fall back gracefully on less capable forms of
connectivity. There are various alternatives that can be used
depending on the application. For example, connected vehicles are
likely to use a variety of modes of connectivity, including 5G, 4G and
peer-to-peer connections. There are various alternatives available
for IoT applications under 4G, such as NB-IoT and LTE-M. Because of
indoor coverage issues, RLANs will need to work together with
public mobile networks. In summary, we are very likely to see a
mixed connectivity environment, with various forms of
connectivity being integrated and 4G and 5G coexisting.
Difficulties of Furthermore, the 5G standards do not define a single service with
defining 5G coverage fixed characteristics, but rather an envelope of performance
characteristics, where there are trade-offs between different
features. Therefore, 5G coverage is not simply a matter of peak data
rates. For example, 5G could be rolled out in rural areas for mMTC
applications for agriculture, yet have relatively low data rates
compared with mobile broadband (even under 4G). A further
example is edge computing. This might be available in certain
124
European Commission, “5G for Europe Action Plan”. Available at:
https://ec.europa.eu/digital-single-market/en/5g-europe-action-plan
92
Regulatory options
locations, but probably not everywhere with a 5G radio layer. Not all
of the IMT-2020 targeted capabilities may be available in a given
location.
Therefore, it makes little sense to draw a binary distinction between
5G coverage being present or absent at a particular location. Rather
there is a set of capabilities, some of which, but not all of which,
might be available. This means that coverage obligations cannot
simply be specified in terms of ‘5G’ presence, but rather need to be
defined in terms of the availability of certain technical capabilities.
Service obligations Typical coverage obligations associated with spectrum have
might be met with a somewhat changed over the last decade or so; from a general
mix of 4G and 5G
obligation to cover a certain proportion of population to
increasingly specific interventions to provide defined services in
certain locations (e.g. a target data speed in certain ‘not-spots’).
However, if the required capability is data speed, then it is likely that
for at least the medium term that requirement could be met most
cost effectively with a mix of 4G and 5G deployment. Again, this
highlights new difficulties with a ‘5G’ specific coverage obligation
and the danger of an excessive focus on achieving 5G roll-out.
Importance of fall Given the potential for mMTC applications in rural areas (e.g.
back connectivity precision agriculture), it is important that there are stop-gap
options for incentive
measures to allow such applications to develop independently of
to develop new
applications and 5G coverage. Applications and services are unlikely to be successful
services in the marketplace if many potential customers are without the
minimum levels of connectivity on public networks needed to
support those applications and services. Whilst one regulatory
response might be to try to provide a roadmap for 5G roll-out (and
use regulatory measures to enforce this), this would still leave
developers of such applications and services in a risky situation if
coverage did not develop as expected. Therefore, we consider it is
likely that developers may address these risks themselves by using
alternative forms of connectivity (e.g. 4G or hub and spoke
arrangements for sensor networks); this provides a fall back so that
new applications and services can be marketed to a wide base of
potential customers. Satellite connectivity may be important in the
short term in providing a basic level of connectivity as a backstop to
allow IoT applications with moderate data rate requirements to
operate in remote areas.
Linkage between In rural areas, availability of high-quality backhaul – particularly
rural backhaul and fibre – is often mentioned as a constraint on 5G deployment.
fibre broadband
However, this is not a ‘5G’ specific problem, as it might also be
interventions
relevant for some 4G sites as data traffic grows.
The question of backhaul for rural cell sites (regardless of
technology) is likely to be closely connected with roll-out of high-
speed fixed broadband to rural areas. Where Member States have
intervention plans to develop deployment of fibre networks to rural
areas, it is likely that these can also provide backhaul connectivity
for cell sites.
93
Regulatory options
7.4 Competitive issues
Private networks and non-traditional entry

Benefits of private We have discussed above the possibility of self-supplied private 5G
networks and non- networks within indoor spaces such as factories. Also, it is possible
traditional entry
that new parties might enter and deploy dense small cells in certain
locations to provide wholesale services to other network operators.
These developments are pro-competitive. The possibility of self-
supply for certain vertical industries is a direct alternative to using a
traditional MNO and so may enhance price competition. Entry of
wholesale cell densification services does not increase the intensity
of (retail) price competition, but nevertheless it puts competitive
pressure on MNOs to rollout 5G networks. There may be physical
constraints on the number of separate small-cell networks that can
be built at certain sites (in some cases possibly only one operator
could be present) creating incentives to be the first there;
competition may be enhanced by the possibility of a whole cell
densifier securing such a site first.
We have discussed above that the spectrum licensing
environment should be friendly to both private networks and
non-traditional entrants because of these competitive benefits.
Supply-side complementarity between fixed and mobile

Fixed networks have Fixed network infrastructure may become important in delivering
a role in providing in- 5G services within buildings. Achieving indoor coverage may
building coverage
become increasingly challenging due to tighter building standards
aimed to reduce energy consumption. It may be unrealistic to
expect public mobile networks to penetrate highly insulated
buildings; using metal foils to reflect heat will also greatly attenuate
radio signals. Therefore, it may be necessary to rely on small cells
within buildings, repeaters or integration with RLANs to deliver a
seamless service.
It will be possible for a mobile-only operator to make use of
available RLANs within a building and integrate them into its
service. 5G will help in delivering a coherent user experience.
Nevertheless, mobile operators with fixed assets may have certain
advantages. For example, it may be possible to use routers supplied
for fixed broadband services to act as 5G femtocells.
Fixed and mobile The increased importance of small cells and the problems of in-
may be supply-side building coverage are likely to create a degree of complementarity
complements as well
between fixed and mobile services on the supply-side (in the sense
as demand-side
substitutes that there may advantages in supplying both). At the same time, on
the demand side substitution of fixed and mobile services is likely to
94
Regulatory options
increase; falling costs of mobile network capacity (both due to

further 4G deployment and then 5G) allow larger data bundles to be
offered. Therefore, there is some degree of tension: whilst
consumers might increasingly want to go mobile-only, there may
also be greater need for fixed services to provide in-building
coverage at high data rates, even though the same terminal device
might be used both in the home and outside.
Increasing The use of 5G technology to deliver FWA services will create a
competition and measure of convergence. MNOs may offer FWA-type services using
possible
mobile network given falling unit capacity costs and the availability
consolidation
between FWA and of 700 MHz spectrum to provide a coverage layer. Indeed, this trend
mobile can be is already happening with 4G as network costs fall and there is
expected surplus capacity at lightly loaded rural cells. Given the potential for
scale economies between FWA and 5G mobile, it is possible that
there might be some pressure for consolidation (at least at the
network level).
Infrastructure sharing and upstream concentration

Infrastructure Infrastructure sharing avoids cost duplication, but may in some
sharing may be cases raise concerns that it could be used as a device to coordinate
essential
retail pricing and inhibit retail competition. For example, a common
network might raise particular concerns about the usage-based
wholesale charges having the effect of coordinating retail pricing
amongst network sharers. Where obligations to supply MVNOs have
been imposed as merger undertakings (e.g. in the Irish O2/H3G
merger) capacity-based wholesale pricing has been required for
just this reason. However, infrastructure sharing is likely to be both
necessary and commonplace for 5G deployment through small
cells. In many cases, it may simply be physically infeasible to deploy
multiple RANs at certain locations.
Certain forms of Therefore, infrastructure sharing is likely to extend beyond
sharing collocation of equipment and tower sharing, with a shared RAN or
arrangements could
with wholesale services being provided by neutral hosts (who might
affect retail
competition be MNOs or other parties). Such arrangement should be beneficial
to 5G roll-out. The likely scenario is that there will be different
sharing arrangements at different locations depending on exactly
which operators are present. However, regulators also need to be
alert to the possibility that if small cells grow in importance and
account for a large share of network traffic, there is the possibility of
concentration upstream onto a single common infrastructure. If this
occurs, certain forms of wholesale arrangements (based on
charging traffic rather than allocated capacity) could then be used
to soften competition in retail markets.
95
Regulatory options
Competition Given the agility of 5G networks to reconfigure network slices and

assessments may adjust the amount of capacity allocated to a wholesale user rapidly,
need to focus more
this distinction between usage-based wholesale charges (which
on physical networks
rather than virtual might have retail competitive effects) and capacity-based charges
ones (which can essentially be treated as fixed costs for the purposes of
retail prices) may become blurred. This may make it more important
to consider the number of competing physical networks when
conducting competition assessments, rather than assuming that
virtual network operators make a similar contribution to
competition as actual networks. This observation is reinforced by
the observation that MVNOs may be able to offer services that are
differentiated in quality relative to the host network through
network slicing.
7.5 Backhaul for 5G cell sites

We have seen that many potential use cases for 5G depend not just
on urban coverage, but more on expectations about rural coverage,
or at least coverage in corridors such as major roads. At present,
backhaul in rural areas at sufficient bandwidth may be limited or
expensive.
Rural backhaul may However, what matters is not so much the current situation, but
be helped by that appropriate backhaul is available on a timescale to fit with
national broadband
likely rural deployment plans. We note that a number of Member
interventions
States are actively intervening in the provision of high-speed
broadband in rural areas through state aid. Given that these
interventions typically promote fibre deployment, there would
seem to be a natural synergy between these interventions and
making fibre available for rural cell sites.
Physical Physical infrastructure access (to ducts and poles) is likely to assist in
infrastructure access deployment of fibre backhaul. However, the impact of physical
might help, but less
infrastructure access is likely to be primarily felt in urban areas
so in rural areas
where ducts are already available. Some rural deployments may not
have existing infrastructure at all to be shared to provide fibre
backhaul. Therefore, it is important to recognise that physical
infrastructure access is unlikely to be a particularly effective
instrument for achieving rural fibre deployment.
7.6 Edge computing

We have not found any live use case that is particularly likely to
make use of edge computing. Nevertheless, in the longer term it is
possible to imagine certain scenarios in which edge computing
could become important.
In particular, augmented reality applications on smartphones could
prove important as a means to overlay information from databases
96
Regulatory options
on the immediate world around us. This could have industrial or

commercial applications (for example, allowing a service engineer
to identify a part by pointing a camera at it).
Many of these applications will probably be able to cope with
latency and use cloud computing resources available on the general
internet. However, in some cases the interactivity associated with
AR might require lower latency. There might also be a need to off-
load extensive computation from handsets with limited computing
power and battery life to servers. Therefore, it is possible that edge
computing could be important for certain niche applications.
Open standards for Suppliers of edge computing are limited to the available network
edge computing are operators at a given location. Therefore, from the perspective of a
likely to be important
customer, edge computing will come bundled with network
to protect
competition connectivity. Open standards for edge computing may be
important to ensure that customers (especially commercial
customers) do not become locked in to a particular provider.
7.7 Net neutrality

The rules regarding treatment of traffic in the EU is contained within
EU Regulation 2015/2120125 (commonly referred to as the ‘Net
Neutrality’ rules) and the BEREC Guidelines126. The rules are directly
applicable across all 28 Member States. Article 3 of the regulations is
intended for ‘Safeguarding of open internet access’:
• Article 3(1) sets out the rights of end-users of internet access
services (IAS)127 in terms of accessing and distributing
information and content;
• Article 3(2) prohibits contractual conditions which limit the
exercise of the end-user rights set out in Article 3(1);
• Article 3(3) regulates ISPs’ traffic management practices,
setting a requirement that ISPs should treat all data traffic
equally (outlining some exceptions);
125
REGULATION (EU) 2015/2120 OF THE EUROPEAN PARLIAMENT AND OF THE
COUNCIL of 25 November 2015 laying down measures concerning open internet
access and amending Directive 2002/22/EC on universal service and users’ rights
relating to electronic communications networks and services and Regulation (EU)
No 531/2012 on roaming on public mobile communications networks within the
Union
126
BEREC Guidelines on the Implementation by National Regulators of European
Net Neutrality Rules, BoR(16) 127, August 2016
127
An ‘Internet Access Service’ is “a publicly available electronic communications
service that provides access to the internet, and thereby connectivity to virtually all end
points of the internet, irrespective of the network technology and terminal equipment
used.” Article 2(2)
97
Regulatory options
• Article 3(4) sets out the conditions under which traffic

management measures may entail processing of personal
data; and
• Article 3(5) sets out that providers of electronic
communications (including providers of Internet Access
Service and providers of content, applications or services
(CAPs)) could provide services other than IAS which may be
designed to meet requirements for a specific level of
quality.
Article 3(3) considering traffic management practices gives relevant
rules for prioritising categories of traffic within the IAS. This is
relevant for certain prospective 5G services. Article 3(5) can also be
considered of particular importance to the 5G debate given that the
delivery of these ‘specialised services’128 may be over network slices
which might allow operators to serve particular needs (e.g. low
latency) to some users separately from the general IAS offering. The
BEREC Guidelines state that, “network-slicing in 5G networks may
be used to deliver specialised services”.129
Under the Net Neutrality rules, specialised services may be offered
where the optimisation is objectively necessary to meet the
requirements of the services. In addition, as specified in Article 3(5)
and Recital 17, such specialised services must satisfy three
conditions:
• “the network capacity is sufficient to provide the specialised
service in addition to any IAS provided;
• specialised services are not usable or offered as a replacement
for IAS;
• specialised services are not to the detriment of the availability
or general quality of the IAS for end-users.”
It will be down to regulators to determine whether the
requirements of a particular case objectively require a specialised
service and whether any of the above conditions are met. The
BEREC Guidelines recommend the information and methods that
NRAs can use to assess traffic management practices and
specialised services when ISPs ask them to provide regulatory clarity
on specific questions concerning implementing new technologies
and services.
128
Specialised services are defined in the first subparagraph of Article 3 (5) as:
“services other than IAS services; they are optimized for specific content, applications or
services, or a combination thereof; the optimization is objectively necessary in order to
meet requirements for a specific level of quality.”
129
Footnote26 of BEREC Guidelines on the Implementation by National Regulators
of European Net Neutrality Rules, BoR(16) 127, August 2016.
98
Regulatory options
7.8 Sectoral regulation and quality of service

Telecoms NRA could In a number of cases, we noted the sectoral regulation (for example,
have a role in aviation regulations for drones) with a predominant factor in
assessing network
determining when, and if, a particular use case might take off.
reliability claims that
affect sectoral Sectoral regulations are often related to reliability issues. For
regulation example, a safety assessment on a 5G-controlled drone needs to
consider underlying network reliability. Similar issues may arise in
medical applications and where 5G is used for monitoring and
control in utility networks.
Sectoral regulators are less well placed to make assessments of
network reliability than specialist telecoms regulators. There is also
a need to ensure that assessments of reliability and security are
broadly consistent across different applications of 5G to different
sectors. Whilst 5G has the capability to increase network reliability, it
is important that the claims are carefully understood and assessed.
For example, even if terminal devices can be simultaneously
connected to multiple cell sites, the reliability improvement
depends on whether there are correlated risks of failure at those
sites.
99
Conclusions
8 Conclusions
What is 5G?
• Whilst 5G may eventually be transformative for some sectors, in the short to
medium term these developments are more likely to be evolutionary than
revolutionary.
• 5G brings enhancements over 4G, including high speeds, low latencies,
enhanced reliability, lower power consumption and greater terminal device
densities. There is an envelope of network characteristics set out in the IMT-
2020 standard within which trade-offs can be struck.
• Network slicing can be used to create virtual networks with particular
performance characteristics by prioritising certain features. This will allow
differentiated services targeting different types of users.
What will drive deployment?
• In the near term, eMBB is likely to be the main driver of 5G deployment. 5G will
help network operators control the costs of growing data traffic due to MBB
and to maintain service quality at competitive levels.
• In the longer run, there may be incremental revenues from new services
enabled by 5G. However, there is no killer application. Rather, we expect there
to be a wide variety of different niche applications. This situation arises
because most commercial uses will be designed for a mixed connectivity
environment of 4G, 5G and other networks and with no certainty about the
dynamics for 5G coverage.
• Industries using 5G as inputs – so-called ‘verticals’ – are likely to have specialist
requirements. This may create opportunities for intermediaries who know
those industries well to assemble connectivity services targeted to their needs.
Connectivity services might involve orchestrating various networks and virtual
networks.
• Another role for intermediaries may result from the need for trans-national
connectivity services to support products and services across the EU. We
already see this in connectivity for cars for infotainment services.
• There is a danger that pricing structures might impede take-up if they
excessively focus on per-connection or per-device charging models. This could
cause some users to inefficiently substitute to alternative technologies.
Private networking
• There is a role for private deployments of 5G within spaces such as factories or
warehouses. Private networks should be pro-competitive and help to drive
coverage of public 5G networks.
• More clarity about which bands to use for private networks would be helpful
to avoid uncoordinated approaches across Member States.
• A secondary spectrum licensing model might be appropriate for private
networks. Unlicensed spectrum with licence licensing might run too greater a
risk of reliability problems for typical use cases (e.g. industrial control).
100
Conclusions
Consequences of small cells

• Small cells are needed to achieve the high data rates needed for 5G and to
exploit mmWave bands.
• Small cells put power into the hands of site owners. Pivotal site owners may be
able to extract rents.
• Restrictive or exclusive access arrangements might in some cases increase
pivotal site owners’ revenue. Obligations on network operators to make
corresponding wholesale access services available to others can remove any
competitive distortion that might result downstream in retail markets, in effect
imposing a neutral host. However, this does not entirely remove the ability of a
pivotal site owner to extract rents.
• In response to this problem, Governments should seek to increase the supply
of sites. Planning rules should not be excessively tight.
• Where public bodies control pivotal sites (possibly in urban areas), there is a
danger that they exploit their power to maximise commercial returns. This
needs to be avoided.
• Fixed operators might have certain advantages in deploying small cells, for
example bundled in consumer routers to support FTTP services.
Backhaul availability
• In rural areas, backhaul is needed for 5G cell sites.
• However, state aid schemes for national high-speed broadband may already
address this issue by encouraging fibre deployment that could be used for
dual purpose.
Infrastructure sharing
• 5G is likely to lead to much more infrastructure sharing, due to the use of small
cells. Sometimes there may be physical limits on how many distinct networks
can be accommodated at a site.
• Charging models for shared infrastructure should ideally be capacity-based to
ensure that there are no retail competitive effects. However, there is a question
mark over how effective this can be in protecting competition when the agility
of 5G networks might allow rapid reallocation of capacity between sharers.
• Regulators need to be alert to excessive concentration due to infrastructure
sharing.
Edge computing
• Edge computing might prove useful for augmented reality or other
applications, but it is too early to tell.
• Open standards for edge computing are important so that consumers do not
become locked in to a particular network.
101
Conclusions
Annex A Glossary
3GPP 3rd Generation Partnership Project
4K 4000 pixels resolution
5G-IA 5G Infrastructure Association
5G-PPP 5G Infrastructure Public Private Partnership
5GAP 5G Action Plan
5GAA 5G Automobile Association
8K 8000 pixels resolution
AR Augmented Reality
ARPU Average revenue per user
B2B Business to business
B2C Business to consumer
BEREC Body of European Regulators for Electronic Communications
BNetzA German telecoms regulation
C-V2X Cellular vehicular-to-everything communication
CAGR Compound annual growth rate
CEPT European Conference of Postal and Telecommunications
Administrations
CFI Call for input
CPS Cyber physical systems
DSCR Dedicated short-range communications
EC European Commission
ECC Electronic Communications Code
eMBB Enhance mobile broadband
EU European Union
FWA Fixed Wireless Access
HD High definition
HGV Heavy goods vehicle
I4MS Innovation for Manufacturing SMEs
IAS Internet access services
ICT Information and communications technology
IEEE Institute of Electrical and Electronics Engineers
IMT International Mobile telecommunications
102
Conclusions
IoT Internet of Things

IP Internet Protocol
IPTV Internet Protocol television
ITS Intelligent Transport Systems
ITU International Telecommunication Union
ITU-R International Telecommunication Union Radiocommunication Sector
LTE Long Term Evolution, a 4G mobile communications standard
LTE-A LTE Advanced Major enhancement of the LTE standard
LTE-M Long Term Evolution for Machines
M2M Machine to machine type communications
MIMO Multiple-input and multiple-output
mMTC Massive machine type communications
mmWAVE Millimetre wave spectrum in bands above 30 GHz
MNO Mobile network operators
MVNO Mobile virtual network operators
NB-IoT Narrowband Internet of Things
NFV Network function virtualisation
NGMN Next Generation Mobile Network
NR New radio air interface developed for 5G
NSA Non-Standalone
NRA National Regulatory Authority
OEM Original equipment manufacturer
OTT Over The Top, communication services delivered over IP
PPDR Public Protection and Disaster Relief
QoE Quality of Experience
QoS Quality of Service
RAN Radio Access network
RLAN Radio Local Access Network (also known as wireless LAN)
RoI Return on Investment
RSPG Radio Spectrum Policy Group
RTR Austrian Telecoms Regulator
SCM Supply chain management
SDN Software-defined networking, a
SME Small to medium enterprises
UAV Unmanned aerial vehicle
103
Conclusions
UHDTV Ultra High Definition TV

URLLC Ultra reliable low latency communication
V2I Vehicle to infrastructure
V2N Vehicle to network
V2P Vehicle to pedestrian
V2V Vehicle to vehicle
V2X Vehicle to everything
VPN Virtual private network
VR Virtual reality
WLAN Wireless Local Access Network
WRC World Radiocommunication Conference
104
Conclusions
Annex B Automotive case study
Within the timeframe of this study, we might expect to see 5G being used to support two main
elements of the “connected cars” proposition:
• improvements to the provision of infotainment services to the vehicle, ranging from
provision of ‘in-car Wi-Fi’ for video and audio streaming, to real-time traffic updates, e-
call services etc., and also to provide remote monitoring data back to the manufacturer
regarding performance of the car and other telematics data;
• vehicle-to-everything communication for remote sensing, safety and control of cars
(e.g. assisted driving and collision avoidance) laying the foundations for realising a
longer-term vision of fully autonomous vehicles.
Basic requirements
For infotainment services, cars need to be connected to the ‘internet’, which will require a
cellular modem that allows data to be sent and received over a public mobile network.
For collision avoidance type services for assisted driving today (with a view to fully autonomous
vehicles in future) at the most basic level, vehicles need to be able to communicate with each
other (vehicle-to-vehicle communication) sharing information such as position, speed, etc. and
need to communicate this to other cars in real time. Vehicles may also need to communicate
with infrastructure (vehicle-to-infrastructure communication), enabling vehicles to receive and
send information to e.g. traffic lights, roadside signs or traffic control centres; and with other
road users such as pedestrians or cyclists etc. Depending on the type of data being transmitted
and the need for processing information may also need to be exchanged with a backend server
(e.g. from a vehicle manufacturer or other mobility service providers) or with the Internet (V2N),
which might rely on cellular data connections. The term Vehicle-to-Everything (V2X) is typically
used to refer to all these types of vehicular communication.130
Can these requirements be met by existing technologies?
Basic infotainment requirements can already be met using 4G LTE services with some
manufacturers already providing such functionality in their cars.131 However, as the data
requirements of services provided over these networks increase and the number of connected
cars on the roads increase, 4G would likely not be able to provide a sufficient quality of service.
As described in the sector overview, V2V communication between vehicles is possible and can
already be achieved using alternative technologies based on short-range communication.
However, to the extent that vehicle-to-everything communications rely on vehicle-to-network
130
As noted by 5G-PPP: “All these V2X use cases rely on the principle that connected
vehicles periodically provide either status information (e.g., position, speed,
acceleration, etc.etc.) or event information (e.g., traffic jam, icy road, fog, etc.etc.). This
information is usually packed into stateless, individual messages or probes which are
either locally disseminated to neighboring vehicles, or sent to a central point (base
station, backend) where it can be aggregated and then again disseminated to other
vehicles to make use of it.” 5G-PPP White Paper on Automotive Vehicle Sectors
131
http://www.wired.co.uk/article/ford-in-car-wi-fi-modem-vodafone-europe
105
Conclusions
communications, this will take place via mobile networks and part of the Cellular-V2X
solution.132
The global cellular standard is now LTE-4G and the current C-V2X is based on 3GPP release 14
specifications, however there are already plans underway to accelerate support from 5G in the
development of the “R16 5G V2X” standard based on 3GPP Release 16 5G standards, and 5G is
expected to bring improvements.
What enhanced capabilities does 5G offer?
For infotainment services, 5G will provide improved data rates and improved mobility for
vehicles travelling at speed, thus widening the scope for more data rich services to be provided.
For assisted driving, the 5GAA consider that “…that 5G will be the ultimate platform to enable C-
ITS and the provision of V2X. 5G will be able to better carry mission-critical communications for safer
driving and further support enhanced V2X communications and connected mobility solutions.”133
Furthermore, “The C-V2X technology is already available but to support the autonomous vehicles of
tomorrow, the technology must evolve to meet more demanding safety requirements. 5G will
facilitate this evolution. Its extreme throughput, low latency, and enhanced reliability will allow
vehicles to share rich, real-time data, supporting fully autonomous driving experiences.”134
5G offers a further (and improved) option for C-V2X communication, and will be more ‘future
proof’ in the shift towards fully autonomous driving in future. For example, “[a]n application such
as controlled fleet driving will require an ultra-low end-to-end latency for some warning signals, and
higher data rates to share video information between cars and infrastructure. 5G should provide the
high reliability, low latency, and high scalability required in this space”.135
In this way, 5G will most likely be an evolution of already available communication technologies
like LTE V2X and IEEE 802.11p but be capable of providing improvements to capabilities and be
capable of delivering further capabilities that will to enable the future V2X use cases136
What are the technical requirements for these enhanced capabilities?
The expected technical capabilities that this use case is expected to require are summarised in
the table below.
Data Rate Latency Reliability Mobility Device Density Position Network

Accuracy Slicing
> 10 Mbps < 5 ms 99.999 % > 200 km/h 10000 /km2 30 cm Preferable
Source: DotEcon and Axon based on information from 3GPP
What are the spectrum requirements for provision of these services?

For infotainment services, it is likely that these will rely on the evolution of public cellular
132
The 5G-AA defines C-V2X as: “…a technology developed by the Third Generation
Partnership Project (3GPP) to deliver V2X services, using two modes of communication:
a direct vehicle-to-vehicle mode (called ‘PC5’ in 3GPP specifications) and a network
communications interface (called ‘Uu’ in 3GPP specifications) for vehicle-to-network
(V2N) communication via existing mobile networks.”
133
http://5gaa.org/about-5gaa/vision-mission/
134
http://5gaa.org/5g-technology/paving-the-way/
135
136
5G-PPP White Paper on Automotive Vehicle Sectors
106
Conclusions
networks as they shift from 4G to 5G. The expectations imply that MNOs would hold the
spectrum rights (in line with the typical model used for 3G and 4G deployment) operated
through exclusive licenses.
By design the specification for C-V2X means that safety critical features associated with device-
to-device or V2V communications will not be reliant on cellular network coverage (as the risk of
drop-outs, no coverage or poor network performance cannot be afforded).
The V2V communication currently requires regionally harmonised ITS spectrum. ITS applications
are intended to be operated in the 5.9 GHz band137, which is allocated for safety and traffic
efficiency. Qualcomm suggests that at least 70 MHz of spectrum is recommended to support
technology and use case evolution of V2X with 30 MHz dedicated spectrum to accommodate
existing/emerging use cases, 20 MHz primary spectrum for future intelligent transportation use
cases or technology migration, and 20 MHz of shared spectrum to be shared with unlicensed
spectrum access technologies138.
For C-V2X, the V2N communications will rely on cellular networks, given the large area over
which 5G would have to be available suggesting that this would rely on public 5G networks
(with good coverage and very high SLGs) rather than a private network. Therefore this would
typically require the involvement of a Mobile Network Operator (MNO).139
Further business model considerations
Once cars are “connected” there will be scope for a large amount of data to be collected on the
car’s movements, locations and possible sensor data (such as acceleration and braking). This
could allow the development of some business models already emerging in the insurance
industry. For example, vehicle insurance premiums are traditionally calculated through risk
algorithms for specific groups of drivers. However, this is not necessarily anything new or
specific to 5G as the industry has already undertaken telematics trials of usage-based insurance
to observer how customers drive and more accurately assess the actual risk posed by drivers
(through pay-as-you-drive or pay-how-you-drive), and some companies are already offering
such deals through simple smartphone apps. Telematics could also detect fraudulent claims
made by customers and, through systems such as eCall, notify insurers of accidents.
Furthermore, once cars are connected: “Automotive manufacturers will be able to use 5G networks
as a platform to open up new revenue streams and business models such as charging for real-time
in-car entertainment, basing rental charges on driving behaviour and route selection, or sale of road
mapping data to third-party organisations. Automotive manufacturers have no experience of
building nationwide communications infrastructure. They can avoid the need to build their own
networks, or to acquire complex skills, by buying managed 5G network services from operators.”140
137
RSPG, Opinion on Spectrum Aspects of Intelligent Transport Systems,
https://circabc.europa.eu/sd/a/b30590d7-5190-480b-b1d1-def24719e061/RSPG17-
008-Final_opinion_ITS.pdf
138
Qualcomm, Leading the World to 5G: Cellular Vehicle-to-Everything (C-V2X)
technologies, June 2016
139
http://5gaa.org/5g-technology/c-v2x/
140
Huawei, “5G Opening up New Business Opportunities. Available at:
http://www.huawei.com/minisite/hwmbbf16/insights/5g_opening_up_new_busin
ess_opportunities_en.pdf
107
Conclusions
Annex C Media and entertainment

case studies
Enhanced mobile broadband for live events case study

This use case is concerned with being able to support the increased data requirements that
result from expanding data usage at large events. For example, the demand for connectivity
may increase due to services being offered to provide an improved experience for
consumers using real-time virtual/augmented reality. These could integrate physical and
virtual information such as scores and information on athletes or musicians. Real-time
services could include replay, choosing a specific camera or language during the event,
watching high definition (HD) playback video, sharing live video, or posting HD photos to
social networks. Visitors could bet, buy merchandise and order refreshments.141142143144 These
services might be made available to consumers through their own mobile devices (i.e. via an
app) or might be provided over devices specific to the event (e.g. a virtual reality headset).
Basic requirements
On-site live event experiences will involve significant demand for downlink and uplink, and a
very high density of devices, with several thousands connected simultaneously in a single
stadium or small area. Next Generation Mobile Networks *NGMN) estimate that a data
volume density of up to at least 0.75 Tbps will be required for a standard stadium in the
2020s145. In order for consumers to be able to make the most of the services listed above, and
to be able to satisfy their demand for data intensive activities at events where there are a
significant number of devices (so that cumulative data requirements may be very large), the
basic requirement is for increased capacity of mobile networks, in a particular location and
improved mobile data services.
Can these requirements be met by existing technologies?
LTE-advanced allows for MIMO techniques and channel carrier-aggregation that can boost
speed and capacity of networks. Increasing the number of cell-sites and deploying small cells
to increase network density is also possible using 4G technologies, and to ‘offload’ traffic
from cellular infrastructure to ‘Wi-Fi' services running off fixed infrastructure. Such
technologies are currently used to temporarily boost coverage and capacity at festivals and
other large events, being deployed in areas where there is a very high likelihood of
significant demand for mobile data at a particular time. However, these solutions may not be
sufficient (and/or cost effective) to meet the increased demand for data at events if this
continues to grow.
141
142
Ericsson, Opportunities in 5G: The View from Eight Industries
143
5G PPP, 5G and Media & Entertainment, 19th January 2016
144
Nokia, Translating 5G use cases into viable business cases, White Paper: Financial
analysis of six 5G business cases to support communication service provider
investment decisions, April 2017
145
108
Conclusions
For example, based on the NGMN estimates above, the 5G-PPP notes that “this very
challenging target, which is not reachable with 4G and its evolutions, will require a seamless
integration of various innovative technologies including broadcast and millimetre waves.”146
What enhanced capabilities does 5G offer?
5G promises to provide a significant improvement in cell capacity and boost data rates,
which will be ideal to accommodate the increasing traffic demands and device density at
large events. This is expected to allow for improved performance and potentially, more cost
efficiently than could be done with existing technologies. For example, Nokia assessed the
relative costs of deploying connectivity to support many people in a stadium streaming HD
video (as part of an immersive experience) and reported the following key findings:
“Capacity: 5G provides up to 40 times more capacity in the stadium than 4.5G
Overall cost: The cost of providing the video services is up to 20 times lower with 5G
Penetration: When delivering high definition video, 4.5G achieves a take-rate (the percentage of
the total audience available to the operator based on its market share) of just 2 percent,
compared to almost 30 percent for 5G multi-camera video steaming.” 147
It concludes: “5G is the only feasible solution for this scenario due to the high capacity density
required. Other technologies would require more small cells than is possible when taking
interference restrictions into account. The better performance with 5G is achieved by being able
to use more spectrum, the much higher spectral efficiency achieved by massive MIMO and
advanced techniques in interference mitigation and receiver technology. The findings show that
an operator wanting to provide good video quality would not realistically be able to use LTE as
this would reach too few subscribers and would incur too high a cost. In contrast, 5G can deliver
high definition video simultaneously to a large number of subscribers within the stadium.” 148
What are the technical/network requirements for these enhanced capabilities?
As previously stated, this use case requires a very high data peak rate with on-site live event
experience being the most important driver for uplink and downlink throughput. Live events
also imply a very high device density, with several thousands of devices connected
concurrently in a stadium or venue. This leads to high traffic density, which may not be
properly attained by the current 4G networks and requires uniform integration of different
innovative technologies such as the use of mmWave spectrum bands, massive MIMO and
densification through the deployment of micro-cells. The use of these techniques will reach
their maximum potential with the deployment of 5G networks.
The expected technical capability requirements for this use case are summarised in the table
below.
Data Rate Latency Reliability Device Traffic Service Network

Density Density Deployment Slicing
146
5G-PPP White Paper on Media Entertainment Vertical Sector.
147
Nokia white paper, “The 5G advantage in real network scenarios Techno-
economic simulation results for solid 5G business and technology planning”
148
Nokia white paper, “The 5G advantage in real network scenarios Techno-
economic simulation results for solid 5G business and technology planning”
109
Conclusions
Time
15-30 10 ms 99% 150000 3.75 90 minutes Effective

Mbps /km2 Tbps/km2
What are the spectrum requirements for provision of these services?

This use case is expected to require a mixture of frequency spectrum of low frequency bands
for coverage and data traffic, and high frequency bands (e.g. mmWave spectrum bands) with
large contiguous bandwidth to cope with the increasing traffic demand, including wireless
backhaul solutions. In Europe, 3.4-3.8 GHz and 26 GHz layers are proposed for stadium and
large events149, with 5G trials at the 2020 Euro Championships representing a practical
realisation of the use case.
Who provides the ‘connection’ will depend on the opportunities afforded by spectrum
authorisations, who holds spectrum rights and vice versa. For example, if spectrum is
allocated on a fixed term, exclusive use basis (e.g. as has typically been the case with bands
designated for mobile broadband use in recent years), then the main connectivity solution
will be traditional MNOs deploying 5G networks on these sites.
However, if spectrum is available on an area defined exclusive use licence, then a new third-
party network provider selling wholesale access or even the stadium/event owner could
provide the cellular network solution. Given high throughput and capacity requirements in
localised hot-spot and congested areas, we understand that some stakeholders have stated
a preference for individual exclusive licenses150.
As discussed in Section 7 of this report, there are a number of options that regulators can
choose from to make spectrum available in order to meet such specific needs.
What are the network requirements for provision of these services?
As discussed in the main report, given the need for indoor or localised coverage and the very
high capacity requirements such a use case will need to be supported with deployment of
multiple 5G small cells with seamless hand-over between cells, which might typically have
very short range due to the propagation characteristics of the mmWave spectrum.
As explored in more detail in Section 6 and Section 7 of this report, there may be a number
of considerations that regulators should be aware of when it comes to site access and the
deployment of small cells.
The future of broadcasting case study
149
ITU and Nokia: 5G for people and things- 700Mhz band as key to success for
wide-area 5G services. Available at: https://www.itu.int/en/ITU-D/Regional-
Presence/Europe/Documents/Events/2017/Spectrum%20Management/Ulrich__No
kia__5G_in%20700w.pdf
150
European Commission: Study of spectrum assignment in the European Union, October
2017.
110
Conclusions
The broadcasting sector will face a number of challenges as the demand for (and creation of)
‘immersive media’ increases. This includes the increase in demand for 4k/8k/ultra-high
definition television with high dynamic range and high frame rates, but also ‘object-based
content’ which includes things such as virtual reality, augmented reality and 360 degree
media.
Furthermore, there will continue to be a shift in demand for the seamless integration of
different broadcasting technologies so that consumers can move from viewing content on
one to the next without disruption. Whilst at present, distribution networks might largely be
considered separate and independent (for example, distribution over cellular, satellite,
broadcast airwaves, digital terrestrial television etc.), 5G may facilitate the development and
provision of a single solution that can “exploit delivery modes for unicast, multicast, broadcast
as well as local caching.”151
One particular project currently on-going seeks to “devise, assess and demonstrate a
conceptually novel and forward-looking 5G network architecture for large scale immersive media
delivery”.152 This project is known as 5G-Xcast.
The main objectives of the project are to investigate whether 5G technology could be used
to support a number of different use cases, by considering how one might:
• “design a dynamically adaptable 5G network architecture enabling seamlessly
switching between unicast, multicast and broadcast and exploiting built-in caching
capabilities”
• “develop 5G broadcast and multicast point-to-multipoint capabilities for Media and
Entertainment” (but also with a view to supporting some Automotive, Internet of
Things (IoT) and Public Warning Systems (PWS) applications)
Work Package 3 of the 5G-Xcast project focuses on the Radio Access network for future
broadcast/multicast 5G transmissions and work is currently on going as part of the technical
research project.
However, a key part of the project is also based on designing a content distribution
framework. This Framework would be network/technology agnostic in order to optimise the
use of a range of networks (e.g. 4G, 5G, home broadband via Wi-Fi) to deliver content to end
users and provide a simple user interface for the content provider so they can code
everything in the same way to deliver to people regardless of the distribution method. The
content distribution framework would provide a single interface (or common set of
standards) that a content provider could use to deliver the content using a single set of
standards for any technology.
Whilst there may be a number of ways to meet these goals using existing technologies, there
is potential for 5G to support the media and broadcasting sector, and the 5G-Xcast project is
exploring these issues.
For example, as reported by the 5G-XCast project:
• Audio-visual media services generate very large volumes of data traffic on networks
which is unevenly distributed over time and geographical areas. At the same time,
Quality of Experience (QoE) is strongly dependent on sustained minimum data rates and
low latencies to all regardless of the total number of concurrent users. This is particularly
challenging for very popular live content (e.g. sports) or unpredictable events (e.g.
breaking news) that tend to cause large traffic spikes.
151
5G-Xcast website. Available at: http://5g-xcast.eu/about/
152
5G-Xcast website. Available at: http://5g-xcast.eu/
111
Conclusions
• The increasing bit-rate demands of 4k UHDTV and, in the future 8k UHDTV, and the
emerging new interactive services (e.g. augmented reality, virtual reality and 360º visual
media) will further increase the demand on network capacity and performance.”153
• “None of the existing networks, whether fixed, mobile or broadcast, has the capability to
support this type of future demand on their own due to limitations associated with
capacity, delay and cost of deployment. Furthermore, the fragmented landscape of
protocols and APIs across them severely limits their ability to cooperate in addressing
this demand.”
• “5G-Xcast will develop a solution that targets such limitations and therefore addresses
future demand, based on the key capabilities of 5G that by far exceed those of the legacy
systems.”
However, it must be acknowledged that 5G usage will not be ubiquitous (or at least not for a
very long time), so one cannot rely on 5G for delivering all of these services and other
methods for distribution may be needed. This will be particularly important for public service
broadcasters (who have obligations to provide a service to all).
153
5G-Xcast website. Available at: http://5g-xcast.eu/
112
Conclusions
Annex D Manufacturing case study
Manufacturing case study

As noted in the main report, many of the potential manufacturing use cases generally fall
under the larger umbrella of massive machine type communication or the Internet of Things.
To the extent that this will require improved wireless connections within a factory there are
decisions to be made about how to source this connectivity (and also the charging models).
We consider some of the important concepts with the help of an example of an augmented
reality service.
Augmented reality could be used to support design, maintenance, process manufacturing,
training, assembly and repair by aiding the execution of procedural tasks through
simulations. Augmented reality (AR) will also enable virtual collaboration amongst engineers
in different locations. This will help to improve communications and effective knowledge
sharing across the industry and will take advantage of the enhanced mobile broadband
capabilities being offered by 5G.
For example, as identified by the 5G-PPP, “the application of augmented reality in the plant will
facilitate:
• Augmented-reality support in production and assembly: Precisely positioned picture-in-
picture fade-ins, it shows the operator the next step and helps avoiding misplacement
and unnecessary scrap,
• Augmented-reality support in maintenance and repair: Repair machine without training
due to augmented information and operational guidance.”154
This may, for example, connect the engineer on site with a back-office support centre who
have all of the necessary information at their fingertips, providing guidance along the way.
Similarly, a connection may be made with a large database of necessary information, which
can then be provided to the engineer on-site.
Basic requirements
The use of augmented reality applications in the work place and factory and an increasing
number of wireless, mobile devices that are using such applications will lead to increasing
demand for the transfer of large amounts of data in these environments (for example,
graphical overlays and models, large data sets, etc.). Therefore, high bandwidth
requirements will be particularly important. The high bandwidth requirements will be driven
by both bandwidth per application/device (with video based applications likely to be a
primary example: “in the case of video-controlled maintenance, with real-time augmented
content mixed into the video signal, bandwidth is important.” 155) and a potentially large
number of those devices running simultaneously.156
Furthermore, low-latency will be required in most of the applications falling under this use
case (for example, where information must be fed in real time and overlaid on video on a
154
5G-PPP White Paper, “5G and the Factories of the Future”.
155
156
Qualcomm, “Augmented Reality and Virtual Reality: the First Wave of 5G Killer
Apps”, 1 February 2017.
113
Conclusions
handheld device). High reliability/availability will also be crucial: “high availability is key to
ensure that (emergency) maintenance actions can take place immediately.” 157 Therefore, high
bandwidth, high reliability and low latency are the most important factors for this use case.
Can these requirements be met by existing technologies? And what enhanced capability
does 5G offer?
Although existing technologies may be sufficient for early deployments of this use case, as
the number of devices using these applications and the requirements of these applications
become more complex, 5G will be crucial to fulfil the full set of technical requirements
needed under this use case in terms of high data rates (bandwidth requirements), coverage
reliability and, in some cases, low latency.
What are the technical/network requirements for these enhanced capabilities?
A highly reliable wireless communication, will be important to integrate the applications of
augmented reality into the closed loop control processes, and provide a seamless experience
while using hybrid wireless and wired network technologies. Augmented reality content
such as overlay on live video being displayed on a mobile device, will involve high data
transmission that can be supported by 5G, for example, as noted by 5G-PPP, “In case of video-
controlled maintenance, with real-time augmented content mixed into the video signal,
bandwidth is important”158
Furthermore, 5G will help support the time criticality of interactive AR support in
manufacturing is characterised by low communication latency159. For some applications
latency may be required to be below 1 ms. It is also worth noting that the need for coverage
is also important to ensure that such applications can be used both on-site and in the field.
The network needs to provide a highly heterogeneous multi-connectivity scenario, where
everything is capable of communicating even in harsh industrial environments. It is also
necessary to provide a fast and reliable configuration of QoS and traffic demands, to enable
fast network adaption.
Furthermore, depending on the exact requirements and the level of dedicated
capacity/control over the network, there may be a greater role for key network management
functions that can be facilitated by 5G networking solutions. For example, as noted by 5G-
PPP, “Concepts such as network slicing, software-defined networking and network function
virtualization are key buildings blocks within 5G, and may become cornerstones in tomorrow’s
communication infrastructure of manufacturers.”160
The expected technical capabilities that this use case is expected to require are summarised
in the table below.
Data Rate Latency Reliability Availability Network

Slicing
10-100 Gbps < 7 ms 99 % 99.999 % Preferred
What are the spectrum requirements for the provision of these services?
This use case is expected to require a mixture of frequency spectrum of low bands for both
157
158
159
For non-interactive content, such as streamed 360° video, this does not apply.
160
114
Conclusions
coverage purposes and data traffic (e.g. for engineers that are using AR applications for
maintenance and support on the field), and high bands (e.g. mmWave spectrum bands) for
capacity purposes.
Therefore, on the one hand, this use case will require exploiting higher bands and making
more flexible use of available bandwidth. Furthermore, leveraging higher spectrum bands
(e.g. 28 GHz), not previously used for LTE, with simultaneous connectivity with spectrum
bands below 6 GHz is expected to ensure a seamless, ubiquitous AR user experience.
The licensing scheme will depend on the particularities of the case. For example, in case of
on-site operations within the walls of a single factory, the use of licensed bands would be, in
principle, feasible with no limitation on technologies and application other than
requirements to avoid harmful interference and reduce risk for interference, which will
disrupt the manufacturing operations.
As discussed in Section 7 of this report, there are a number of options that regulators could
choose from to make spectrum available in order to meet such specific needs.
Business model implications
The manufacturing industry players are expected to incur costs in procuring the devices,
software, and training that would help them smoothly integrate AR into the various
manufacturing processes. However, these costs may be recouped through improvements to
efficiency and productivity in training and maintenance activities - for example, support can
be provided more quickly, with off-site specialists communicating with local engineers in
real time, using augmented reality to supervise activities such as maintenance, repairs,
equipment upgrades or training. This will remove the need to wait for specialist advisors to
come on-site and reduce the need for in-depth training for new recruits. In this way, the
introduction of new remote services such as AR could lead to the re-structuring of teams
within the manufacturing environment, with the creation of virtual back office teams of
specialists. “These remote teams may use the data coming from smart devices for preventives
analytics and easy access to work instructions” 161
There may be scope for new intermediaries in this market to develop specific AR solutions
tailored to the factories needs which may be made up of an entire solution (including the
wireless network) or provide the means by which the application can be integrated with
devices and the networking arrangements already in place.
One of the main issues will be around who provides the network. The main cost driver for
connectivity providers will be in providing the necessary infrastructure (e.g. small cells,
power, backhauling). Given the potentially significant costs involved and depending on the
exact demands of the network the manufacturer that wishes to use 5G to support AR
applications in the work place will face a build or buy decision.
If a solution is to be provided that is reliant on public mobile networks, then MNOs will see a
change in their business model through shifting away from the traditional B2C model and
towards a B2B relationship where it will need to offer a bespoke solution to the manufacturer
and set prices in such a way that allows them to get a RoI.
However, depending on the extent to which there are connectivity requirements for other
parts of the factory (such as machine-to-machine communication, remote monitoring
devices, or services relying on mission-critical communications, for example), the
manufacturer may demand some control over the network rather than relying solely on an
external provider or MNO. This will require the factory owner to come up with their own
161
115
Conclusions
networking solution (subject to spectrum licencing conditions).

Many of the issues regarding the sourcing of ‘connectivity’ are similar to the ‘events’ case
study, however the issue of interoperability across different mobile networks is unlikely to be
as relevant given that the network will be used only for private purposes. We discuss these
issues in more detail in Section 6 and 7 of the main report.
116

LTE Parameter Tuning

Uploaded by

Copyright:

Available Formats

LTE Parameter Tuning

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

LTE Parameter Tuning

Uploaded by

Copyright:

Available Formats

How will 5G impact different industries?

How will 5G impact different industries?

What are some of the technical capabilities that differentiate 5G from previous generations?

What are some of the technical capabilities that differentiate 5G from previous generations?

BoR (18) 23

A report by DotEcon Ltd and

7.8 Sectoral regulation and quality of service ......................................................................... 99

Tables & Figures

power wide-area technologies such as LoRa in a hub and

Pricing structures and efficient take-up

2 Background and context: the

Source: DotEcon and Axon

Each transition has been driven by the greater capabilities of each

ahead to identify ways their networks can be readied to meet future

multiple ‘virtual’ networks with differing performance

2.2 Global trends in information and

Table 1: Role of 5G in global telecom trends

Major trend Key role of 5G

Source: DotEcon and Axon

Connectivity: There has been a massive increase in demand for

Proliferation of value-added services and over-the-top (OTT)

networks will be aiming to serve not only individual customers, but

2.3 The roadmap towards 5G

Figure 2: Expected roadmap towards 5G

Vision, Workshop Evaluation

• Uninterrupted 5G coverage in all urban areas and

5G Private Trials, 5G Platforms, 5G for Euro 2020,

2017- 2018 2019 2020 2021 2022 2023

Source: DotEcon and Axon

• Full compliance with ITU´s IMT-2020 requirements is

research projects and cross-project working groups, will help deliver

2.4 Regulatory environment

EU-level identification of spectrum for 5G

technical conditions, as well as sharing conditions for 5G

Support for investment in 5G

authorities, which will be more suitable for 5G deployment. These

Consultations at a national level

recommends studying the possibility of using the 40.5-43.5 GHz (as

expire shortly. However, 10-year licences are proposed – shorter

the deployment of 5G throughout the country, including a tender in

3 Overview of key 5G technologies

u Adoption and utilization across

u Uptake of mobile technologies

u Improving network ULTRA-REALIABLE, LOW-

u Applications requiring high

u Strong security and availability

Source: DotEcon and Axon

To effectively satisfy these use cases, 5G must combine a number of

3.1 Technical capabilities that differentiate 5G

User 10 - 100 Mbit/s 10k - 1million devices/km 2

90% more efficient

350- 500 km/h 0.1– 10 Mbit/s/m 2

IMT-2020 defines an Technical capabilities under IMT-2020 present a significant

Figure 5: Enhancement of key capabilities from IMT-Advanced to IMT-2020

Source: Recommendation ITU-R M.2083-0

Enhanced mobile broadband (emBB)

Ultra- reliable and low latency communications (URL L C )

Massive machine type communication (mMTC )

Source: Recommendation ITU-R M.2083-0

u  Adoption and utilization across

u  Uptake of mobile technologies

u  Improving network ULTRA-REALIABLE, LOW-

u  Applications requiring high

u  Strong security and availability

u  Connected u  Live events u  Logistics u  Drone delivery u  Agriculture u  Smart u  Remote