What are digital twins? Three concrete examples

Over the past few decades, what we expect from our devices has changed dramatically. A phone isn’t just for calls or messages anymore, a car probably knows the way to your destination better than you do, and our industries and cities are getting smarter and more connected by the day, powered by 5G. and the IoT. We see our future being built around us – a future of highly complex networks and interconnected digital ecosystems.

With huge volumes of real-time data continuously generated by every device, managing these networks and ensuring they operate efficiently and sustainably will rely heavily on artificial intelligence (AI) and machine learning (ML) . But training poses a challenge. After all, how do you safely train an ML algorithm to learn and take responsive action in a dynamic live network where downtime is not an option? This is where digital twins come in.

Faithful representation of a real site in a digital twin

What is a digital twin?

A digital twin is essentially a copy – a software representation of all the assets, information and processes present in the real, but cloud-based version. Digital twins open up a virtual world of possibilities – a safe simulated test environment in which you can train and play out “what if” scenarios to your heart’s content (or the training model), without risk to the peer of the real world.

Explore digital twin use cases

Although they may sound like science fiction, digital twins are already being harnessed in commercial solutions, unlocking the potential of AI, data and digitization to enable the mobile networks of the future. Here, we explore three real-world examples of digital twins and learn how this technology opens up new possibilities for optimized, automated, and scalable networks.

Use case 1 – Network Digital Twins: A secure approach to automation for 5G networks

A network digital twin models what we think of as the invisible network: signals, coverage, interference, and traffic behavior, including user mobility across frequency layers. The digital twin ensures a safe approach to optimization, an essential factor when it comes to sensitive parameters, such as radiated power, for example.

In Switzerland, our customer Swisscom operates the highest rated network in the world according to the umlaut international benchmark ranking for 2021. But it is also subject to some of the most stringent radiated power regulations. Without changes to existing infrastructure, regulations would only allow the deployment of a few new low-power 5G sites, resulting in patchy coverage on a new low-band layer that would be used by both 4G and 5G New Radio (NR).

So how could we reduce the transmitted power to make room for the new layer, without compromising coverage or user experience?

We knew the best approach would be to use reinforcement learning (RL) – a machine learning methodology where an agent interacts with the environment by observing its state and taking iterative actions that gradually converge towards a goal at long term. Our long-term goal was to reduce the transmitted power. But we couldn’t allow the agent to mess with the power radiated into the real network, because that could compromise the user experience and violate the very regulations we strive to adhere to.


Report on improving the customer experience with AI

Learn more about reinforcement learning and how AI improves customer experience in a complex 5G world

Download report

Our experts have developed an accurate digital twin of the network that models coverage, interference and traffic behavior, including user mobility across frequency layers, providing a safe environment in which the RL agent can play and learn. . In-depth domain knowledge was crucial in selecting what to model and finding the balance between making the twin accurate and detailed enough, yet simple enough to run in a scalable commercial application.

A safe application to reinforcement learning powered optimization through the use of network digital twins

Figure 1: A safe application to reinforcement learning-based optimization through the use of network digital twins

After thousands of learning cycles, we implemented the final set of recommendations. Almost all cells in the area had their power changed, both up and down, resulting in an overall 20% reduction in transmit power with a base station power consumption of 3 .4% lower. Interestingly, the coverage remained intact and the user experience actually improved, with 5% better download speeds and 30% better upload speeds.

Turns out, in networking, much like conversations in a crowded restaurant, shouting louder will only get you so far — but if everyone lowers their voices, we can hear each other better.



Design and optimization of networks based on machine learning


Use case 2 – Site Digital Twins: a new era for site and equipment management

A site digital twin models the visible network – the towers, equipment, and all other assets included in a physical site. As 5G technology accelerates, we need to ensure we can expand and maintain networks quickly and efficiently. But in reality, on-premises asset lifecycle management is often far from agile. There can be over 20 documents describing what is installed at a single physical site – from CAD designs and images to spreadsheets and product data sheets. This manual documentation makes the process slow and error-prone, and often ends in unnecessary site visits and mast climbs.

To kiss "Digital twins"

Figure 2: Digital twin of the site

Using a digital twin, we are bringing the IKEA kitchen planner to the telecommunications industry, enabling fully digitized design and management of sites and equipment. We have a single digital twin for each site, with an accurate 3D model captured with laser scanners (LiDAR), cameras and drones. The twin includes all the key metadata necessary for effective and efficient lifecycle management, including constraints such as weight, power, and compatibility between components.

We have also developed a library of 40,000 components, with each component available to be easily “dragged and dropped” into place. This reduced design time by 50% and improved maintenance, reducing the number of site visits from one in ten to one in a thousand. This means fewer trips to site and fewer people having to climb masts, for safer, more predictable and more sustainable operations overall.

Use Case 3 – Subscriber Digital Twins: Bringing 5G into the Omniverse

We have a network twin, we have site twins, but there is still a key third dimension missing in our digital twin trifecta – the subscriber. Our research team collaborated with NVIDIA Omniverse to bring CGI technology from games and movies to the telecommunications industry, enabling real-time subscriber modeling using the Unity game engine.

This revolutionary crossover will involve the evolution of internal network models with an accuracy never seen before in real-world measurements. Essentially, we take 3D gaming technology – and its extremely high computational complexity of physically accurate models – as a base, and then we deploy propagation models for 5G on top.



Welcome to the Omniverse


Welcome to the Omniverse: Ericsson’s expertise in radio network simulation meets NVIDIA’s technologies in rendering and collaborative design.

The technology enables modeling of complex urban and interior geometry at high resolution, including bridges, tunnels, foliage and detailed modeling of surface materials that influence radio frequency (RF) propagation, and modeling of user mobility and dynamic scene characteristics such as automobile traffic. It also uses Pixar’s open Universal Scene format, which allows for the reuse of detailed city meshes and geodata, which is sometimes one of the biggest challenges in accurately modeling an environment.

As we know, future networks will only become more complex, so models will need extensive visualization support to be meaningful. NVIDIA Omniverse Create integrates a state-of-the-art ray tracing engine with interactive tools to manipulate and explore complex scenes, allowing us to experiment with Ericsson product placement and explore their impact in real time – a true catalyst for the best performing product development.

The future of 5G is bright and full of virtual worlds full of possibilities.

Learn more

Learn more about AI and reinforcement learning in telecommunications.

Discover how AI is applied to achieve efficiency and performance in networks

Find out what else is possible with smart site engineering from Ericsson.

Discover our collaboration in 5G simulation on the Omniverse platform.

Discover how digital twins are modernizing the oil and gas industry and transforming port operations.

Learn more about the future of digital twins in mobile networks in our blog post.

About Shirley L. Kreger

Check Also

Examples of Good Retail Customer Service

It’s a known fact that retail businesses must provide excellent customer service to be successful. …