Digital design for 5G:

A disruptive use
case-driven approach

White paper

The sheer complexity of 5G networks that address the needs of multiple and highly
specific use cases will call for an innovative approach to network design; one that is driven
by data analysis, automation and real-time simulation.

The new 5G digital design process will also be more inclusive, involving all stakeholders and
ensuring that, as the network design is optimized, the needs of the use case will be met.
 Contents

Executive Summary: Focus on the use case, not the network 3

It’s the end of the road for conventional network design 3
5G will bring a substantial leap in performance … and complexity 3
Use case focus calls for inclusive design 5
Conventional design processes will be overwhelmed 6
A new approach to 5G network planning and design is needed 6
Understanding the use case requirements 8
Creating the network design 9
Fast and accurate network design 9
The technologies to support digital design 9
A practical example of network design for use case requirements 10
Conclusion: The digital transformation of network design 11
Abbreviations 12

2 White paper
Digital design for 5G: A disruptive use case-driven approach
Executive Summary: Focus on the use case, not the network
Until now, telecom network design has been largely static and based on creating a one-size-fits-all
infrastructure. However, the arrival of 5G brings about a new focus on specific use cases and how they
can be supported. A roll out of 5G infrastructure will support a wide range of use cases with varying
performance needs. Network slices will create virtual instances of a network, using the same underlying
hardware, to meet each use case’s needs.
In addition, adding another network generation with new radio spectrum and technology will bring about a
further jump in complexity that conventional design processes will not be able to deal with.
Instead, a network design process that’s dynamic and automated, uses real-time simulation and directly
involves multiple stakeholders, will need to be used. Analyzing huge amounts of data from many sources
to build up a digital twin of the use case and its environment will provide the design parameters for the
network. This is then fine-tuned in an iterative process to produce a final design for a 5G network or a
network slice that meets the precise performance needs of the target use case.
As well as being able to handle the huge complexity of 5G deployments, such a design process will help
communications service providers and enterprises to shorten the time to implement a use case and
ensure a more precise network design that requires minimal post-deployment optimization.

It’s the end of the road for conventional network design

For more than two decades, mobile telecoms has undergone constant and rapid change. As each new
technology generation has been deployed and then evolved to deliver ever-higher performance, network
design has become increasingly intricate and time-consuming.
New network layers have been added on top of existing ones. Radio Access Networks (RANs) have taken
new spectrum into use alongside existing bands, while using a more diverse range of sites, from the
traditional large base macro base station site to the many different types and sizes of small cell access
points.
As more services have been offered, driven by the rise of mobile broadband, core networks have expanded
in complexity and moved wholesale into the cloud.

5G will bring a substantial leap in performance … and complexity

Yet, with the arrival of 5G, network complexity is about to explode beyond anything seen so far.
5G networks are expected to be capable of meeting wildly diverse and demanding requirements. Video
content and the Internet of Things (IoT) will call for 100x more network capacity. Extreme low latency will
be required to meet the demanding performance needs of vehicle communication and remote control.
Distributed cloud computing with processing being performed in the core network and at the network
edge will become the norm. Ultra-reliable and low-latency access will be required by mission-critical
services.
Achieving ultra-low latency depends not only on the radio design, but also the mobile access. Meeting the
end-user requirements of 5G requires an end-to-end network design approach.
New spectrum will be used, with very short wavelength bands being prone to attenuation by the
surrounding environment including foliage and even the physical presence of people. Complex new
antenna technology will be deployed, using technologies such as massive Multiple Input Multiple Output
(MIMO), beam forming and beam steering to ensure coverage.

3 White paper
Digital design for 5G: A disruptive use case-driven approach
The different frequency bands and combinations of bands used will need to be taken into account by
the network design to ensure their optimal use. Similarly, mobility between different beams needs to be
addressed when using beam forming.
Virtualization and Software Defined Networking (SDN) enable network slices to be created to support new
services with different Service Level Agreements (SLAs) and with no disruption to existing services.
With network slicing, Communications Service Providers (CSPs) and enterprises will be able to address the
specific needs of different customer segments. For example, they will connect the factory of the future
and help create a fully automated and flexible production system. In healthcare, hospitals will be able
to arrange remote robotic surgery. City governments will be served for real-time traffic management,
emergency response and video surveillance with customized quality and security requirements. Even within
a city, separate slices may be created for different municipalities and organizations for administrative
reasons, even though they may have technically similar requirements.

Figure 1. 5G network complexity and the need to fulfil extreme performance demands spell the end of the
road for conventional network design

Hypothetical network Today 2020-25

Users 10M people +100M ‘things’
Speed 100 Mbps 100x faster
Latency >>10 ms 10x less
Network service level Best effort for all Committed SLAs
Logical networks 1 Many (slices)
New Services that need
more network design
Smart home Mobile gaming Connected cars Industry 4.0 Drones IoT wearables

4 White paper
Digital design for 5G: A disruptive use case-driven approach
 What is a network slice?
5G network slicing is a new network approach that can provide highly tailored services to specific
customer groups and even individual customers. Virtualization and cloud technologies enable the same
underlying network infrastructure to be used to create virtual network instances with performance
characteristics that fit the needs of new customer groups, something that conventional one-size-fits-
all networks cannot achieve.
These network slices can be set up relatively quickly and managed dynamically to deliver services that
can command premium revenue. There is a large potential base of customers within public safety,
industrial automation, healthcare, Internet of Things (IoT) and many other areas that would value such
services.
Furthermore, slicing allows the same infrastructure to be dynamically shared by different tenants, thus
optimizing the Total Cost of Ownership (TCO).
Network slices are composed of capabilities from multiple network segments from the access to
the core, as well as applications. Only an end-to-end network design approach can ensure that
performance requirements of different use cases can be met with certainty.

Use case focus calls for inclusive design

The design of a 5G network is driven by the needs of the use cases that will be supported. This is a
fundamental shift from conventional design processes for previous generation technologies that focused
on creating a one-size fits-all network.
While 5G will provide general broadband coverage, it is the underlying use cases with their specific
performance characteristics that must be addressed from the outset by the design process. Such use
cases could include automated factories, connected cars, connected healthcare, camera surveillance and
many more.
The performance demands of these use cases will vary tremendously in terms of the capacity, coverage,
throughput and latency required, making the design process highly complex. As well as the hardware and
software infrastructure design encompassing sites, backhaul radio functions and so on, the optimization of
multiple network parameters also needs to be considered for each use case, such as antenna tilt, azimuth,
slice parameterization, capacity and more.
With use cases leading the design process, it becomes vital for the end customer, such as the CSP,
enterprise or authority, and other stakeholders to be involved in the network design process. This inclusive
design approach ensures that not only will the network design enable the use case to meet the customer’s
needs, but that the business case remains valid. For example, if the network design calls for unanticipated
investment in new hardware, the business viability of the use case may need to be revalidated. The
involvement of all parties is vital, especially when a CSP is dealing with different network vendors to create
multiple slices for several use cases.

5 White paper
Digital design for 5G: A disruptive use case-driven approach
Conventional design processes will be overwhelmed
The sheer complexity of 5G network planning and design cannot be resolved by conventional techniques.
5G networks will address a wide scope of use cases with a huge variety of requirements and is likely to
involve the deployment of multiple network slices. It is vital to understand the impact of adding new
network slices; how they may affect the performance of the rest of the network and how the network must
be modified to support the slice, for example by adding capacity and coverage at specific locations.
This complexity is likely to engulf traditional network design techniques that are reactive and based on
a process of design, deploy and verify. Such processes are highly manual, process intensive and lack
flexibility. Once a CSP or enterprise has established the business viability of a use case, it would need to
deploy the network to support it as quickly as possible. A conventional network design process applied to
5G is likely to be lengthy, costly and too slow.
Instead, an approach that involves real-time simulation and dynamic digital design will be needed. Such
a process must be able to handle huge amounts of input data from the network and its surrounding
environment, the specific needs of each use case and the collective knowledge gathered from previous
and existing network deployments.
And unlike previous network generations, which are designed to provide similar services to all end
users, 5G network design will be driven by specific use cases, each with its own capacity, coverage and
performance needs.

A new approach to 5G network planning and design

is needed
Applying machine learning and data analytics makes it possible to digitally model specific use cases as a
basis for a new, more dynamic 5G network design process.
This ‘Digital Twin’ provides the performance characteristics the network must fulfil. These parameters are
used for the end-to-end 5G network design across different domains and simulate its performance. The
simulation is then fed back into the use case model in an iterative process that progressively fine-tunes
and optimizes the network into a final detailed design.
Feedback and interaction between the different components of the design process and with stakeholders
is critical to help ensure the final network will meet all the performance requirements.
Applying a predictive and intuitive design methodology helps in the understanding of the different design
options and their impact on the network’s design, cost and performance.
The overall digital design for 5G process is illustrated in figure 1. The process can be used both for
designing a new 5G network deployment and for the design of network slices to meet the needs of a new
use case that needs to be supported by an existing 5G network.

6 White paper
Digital design for 5G: A disruptive use case-driven approach
 What is a Digital Twin?
A Digital Twin is an exact digital replica of complex physical assets, processes and systems that also
interact with the real system. The twin can be fed data so that it changes in the same way as the real
thing. This then provides a detailed understanding of how the real system is behaving and predict what
it will do next.
When it comes to network planning and design, a Digital Twin goes far beyond current software tools.
Using artificial intelligence, it is more granular, more dynamic and more accurate than traditional
techniques. The Digital Twin looks at everything that affects the network including data about where
people are using the network and how, as well as trends gathered from social media.

Figure 2. A new digital design process for future 5G networks

Ecosystem

CSP

Enterprises Suppliers

Real world Use case Digital world

implementation requirements simulations

Authorities App Developers

Accurate network assessment & implications

Nokia
AI/ML
AVA

7 White paper
Digital design for 5G: A disruptive use case-driven approach
Understanding the use case requirements
The first step is to understand the detail of the use case. This is at the heart of the process and uses
an intuitive tool that analyses the use case environment, such as a factory site or an urban municipality,
enabling it to be simulated digitally and visualized in 3D. Visual modeling simplifies decision-making and
aids discussions between the various stakeholders, including the CSP, enterprise, or authority ((such as a
harbor, airport or transport authority) requiring the use case.
Understanding the use case requirements must consider all available information about the environment
in which it must run. Many different sources are used, including static data like terrain mapping, built
environment databases, end user data and environmental data such as the location of trees and foliage
that can affect radio signal propagation.
Real-time data from the existing networks (non-5G and 5G if deployed) and other data sources can also
be taken into account. The use of real-time data will allow a more accurate assessment of how the final
network would affect the use case and the potential impact on existing services.
Further valuable sources of data that can be used as input for the planning include previous network
deployments, existing use cases and expert network design knowledge. This can include the application of
existing algorithms developed for other use cases, machine learning solutions trained on earlier use cases
and reference use cases.
Insight into potential design constraints and Key Performance Indicators (KPIs) for similar use cases can
also be useful.

Figure 3. A 3D visualization of the use case requirements sits at the heart of the digital design process

8 White paper
Digital design for 5G: A disruptive use case-driven approach
Creating the network design
The network design process takes the use case requirements and applies techniques such as advanced
simulation and Automatic Cell Planning (ACP) to first create a high-level design that determines
site locations. It also encompasses basic dimensioning and configuration and the topology and
parameterization of the network slice to meet performance requirements such as data rates, latency and
so on.
The simulated performance of the high-level design is fed back into the use case requirements in an
iterative process. This aims to optimize both the network design and the use case to create a detailed
design for the network or network slice. The optimization encompasses all aspects of the network,
including the hardware and software design, end-to-end connectivity, final network topology and detailed
parameterization such as antenna tilts and other requirements, such as power supply demand.
The detailed design is used for deployment of the entire new network or the addition of hardware as
required to support a new network slice.

Fast and accurate network design

While it is possible to run use case driven network design today, using conventional techniques is slower
and less efficient.
The 5G digital design process will bring numerous benefits to CSPs, enterprises and other stakeholders. The
final detailed network design can be produced much faster than is possible with conventional techniques. It
is also far more accurate requiring minimal post-deployment optimization. Together these attributes mean
a use case can be brought to market and be operational in very little time for a fast time to market.
Real-world simulation will also provide an accurate calculation of the investment costs involved, enabling
the business case to be understood with a good degree of confidence, facilitating decision making.

The technologies to support digital design

The digital design process for 5G networks is only made possible by some of the latest technologies such
as artificial intelligence, automation and data analytics.
An accurate understanding of use case requirements can only be achieved when detailed data about the
environment and performance requirements are available. The wider the variety of high-quality data, the
better.
Nokia has studied and is developing the required tools and methodologies to gather the necessary data
such as user profile segmentation, user locations in 3D and forecasting of new applications in future use
cases, based on machine learning models that provide the right inputs for a correct design.
A detailed and clear visualization of the use case environment will help all stakeholders to understand the
design parameters more clearly and will be achieved using a 3D environment similar to the latest game
engines, such as the Unity game development platform.
Nokia gathers and applies this wealth of data as part of its AVA solution which also acts as a store for the
learned experience from network designs and deployments globally.
As well as providing an efficient means of delivering services, Nokia AVA is a giant repository of learning.
With the integration of AVA, the 5G digital network design process can take advantage of Nokia’s deep
network design knowledge and understanding of 5G technology to produce detailed performance
predictions.

9 White paper
Digital design for 5G: A disruptive use case-driven approach
 What is Nokia AVA?
Nokia AVA (Analytics, Virtualization, Automation) is a cloud-based open platform that allows experts
worldwide to share their knowledge and best practices and to use leading-edge big data and predictive
technologies to vastly improve network performance.
Nokia AVA enables the deployment of advanced services to CSPs, authorities and enterprises by
integrating three key capabilities:
• Analytics to collect and process huge amounts of data. Machine learning bring numerous benefits,
for example recognizing problems before they can affect network performance or to quickly find
configuration issues that can be corrected automatically
• Virtualized delivery for flexible and fast commissioning of services. Nokia AVA is cloud-based to
flexibly deploy new use cases in days or even hours, instead of the weeks conventionally needed.
• Automation to filter and process data quickly and accurately implement and configure networks to
deliver smoothly running services at all times.

A practical example of network design for use case

requirements
To see how the digital design process is applied in practice, consider a factory manufacturing electronic
equipment. The enterprise wants to improve the productivity of a product’s assembly that requires manual
work.
The aim is to help assemblers improve their technique and alert them immediately should an error occur.
This would allow errors to be corrected during the intricate assembly stage, thus avoiding defective
products being rejected by subsequent quality control and sent back for rework.
The use case requirements are based on real-time video surveillance of the assembly process, which
requires ultra-low latency and high bandwidth, to enable high-quality video to be captured and analyzed
immediately by artificial intelligence.
The use case requirements are initially modeled in 3D showing the location of the assembly bench and its
surrounding environment. This modeling involves the enterprise, the mobile CSP, the 5G network vendor,
the video processing IT supplier and the video application builder.
The use case requirements model allows a high-level network design that places the 5G base station in the
best position to deliver reliable, interference-free 28 GHz band connectivity from the video surveillance
cameras to the local processing data center, deployed to achieve the necessary low latency radio network
topology. The design process is shortened by applying and modifying an existing specification from a
similar Industry 4.0 application suggested by the knowledge database.
Using the high-level design, the stakeholders adjust the use case to improve the performance of the
video application based on the predicted network performance and taking into account a planned future
expansion of the assembly area. Using wireless connectivity instead of fixed provides the flexibility the
factory needs.

10 White paper
Digital design for 5G: A disruptive use case-driven approach
The optimization then produces a final detailed network design including 5G base station and with local
processing to achieve low latency that supports the integrated video analytics application. The use of MEC
together with 5G allows data to be processed close to where it’s needed, dramatically reducing latency.
The video application monitors the video feed and uses machine learning to analyze and immediately
alert the assembly line operator of any inconsistencies, so they can be corrected in real-time, helping to
maintain maximum manufacturing productivity and product quality.

Figure 4. Applying the 5G digital design process to a factory use case

• Ultra low latency

• 100% connectivity

Define the requirements

• Predict radio network

performance with
modified factory
new line up

Use Case simulation

• Update design for

modified factory layout
• New antenna locations
and their configuration

Network design and real world

Conclusion: The digital transformation of network design

5G networks will be able to address the highly specific needs of many different customer segments. A huge
range of use cases can be supported with highly specific performance capabilities such as ultra-low latency
and extreme throughput. 3G and 4G networks were designed with a focus largely on producing a one-size-
fits-all network. But the business viability of 5G will be driven by the use cases it supports. This means 5G
network design must start with the needs of each intended use case.
The Nokia 5G Digital Design process brings a use case driven design approach that addresses the end-to-
end design of 5G networks. The approach will build on digital modeling, intuitive design and agile techniques.
The iterative design concept consists of some key components:
• Understanding the use case requirements
• Real world implementation
• Accurate network assessment and implications
• Digital world simulations
• Artificial intelligence and machine learning to process large amounts of data from various inputs

11 White paper
Digital design for 5G: A disruptive use case-driven approach
Adopting such a use case approach to 5G networks calls for a new digital design process with 3D modeling
and advanced performance simulation at its heart. Conventional design methods will not be able to cope
with the complexity of 5G network design.
The 5G digital design process will be more inclusive, involving all stakeholders in the creation of the use
case and ensuring that as the network design is optimized, the needs of the use case will be met. This will
help to meet stakeholder business needs and prioritize investments, while also achieving a faster time to
market.
The accurate design of the 5G network and associated network slices is critical to ensure that premium
services can be supported properly. Nokia 5G Services provide the expertise and processes to achieve
this goal.

Abbreviations
ACP Automatic Cell Planning
CSP Communications Service Provider
IoT Internet of Things
MEC Multi-access Edge Computing
MIMO Multiple Input Multiple Output
RAN Radio Access Network
SDN Software Defined Networking
SLA Service Level Agreement
TCO Total Cost of Ownership

About Nokia
We create the technology to connect the world. Powered by the research and innovation of Nokia Bell Labs, we serve communications service providers, governments,
large enterprises and consumers, with the industry’s most complete, end-to-end portfolio of products, services and licensing.

From the enabling infrastructure for 5G and the Internet of Things, to emerging applications in digital health, we are shaping the future of technology to transform
the human experience. networks.nokia.com

Nokia is a registered trademark of Nokia Corporation. Other product and company names mentioned herein may be trademarks or trade names of their respective owners.

© 2018 Nokia

Nokia Oyj
Karaportti 3
FI-02610 Espoo, Finland
Tel. +358 (0) 10 44 88 000

Document code: SR1806026163EN (June) CID205313