Seeing Double: How Digital Twins Are Being Used to Improve 5G Network Design

By Dr. Nishita Hathi, Product Director, 5G Network Planning & Design Solutions
Part 2/3 of our 5G network design blog series

5G networks have become so complex that the industry is turning to digital twins to assist with rolling them out. A digital twin, as defined by Gartner, is “a digital representation of a real-world entity or system.” In the telecom world, this real-world entity typically includes a service provider’s network, its broader environment (such as terrain, buildings, vegetation), subscribers, devices, and so on. A digital twin can duplicate the live network environment so that it can be used to quickly and cost-effectively to simulate all types of interactions to determine what will happen in different scenarios, using different 5G features and capabilities

Why is this important? Telecom networks are so large and complex that changes are slow and costly. Innovation can be risky in this environment. Small mistakes can have an outsized impact and can cascade to create network failures and lost revenue. Meanwhile, as the 5G rollout accelerates over the next five years, operators will need to integrate nearly 15 million new 5G sites into their existing network. That is about 50% faster than 4G. By modeling network changes in the digital domain, CSPs can design and rollout new technologies more quickly. They can try out an endless array of new parameters and capabilities much more efficiently and cost-effectively, making the right decisions around network design and rollout from the get-go in just a fraction of the time. (Figure 1, below)

5G Network Design Challenges

There are some unique challenges when designing and rolling out 5G networks. First, there is the enormity of it all. Typically, each new cell site that’s introduced into the network has thousands of parameters. Multiply that by the number of new cell sites, which can be in the thousands, and you get the picture.

Second, network designers must also factor in all the equipment and systems from multiple vendors with various technology configurations. This is especially true in the context of 5G. Each cell site requires time-consuming, careful planning so that it’s optimized for its environment. A team of network designers with decades of experience usually takes this on, but with the number of new 5G sites needed, there simply aren’t enough specialists – or hours in the day – to go around. Besides, network engineering teams already spend too much of their day doing repetitive tasks. What they need are the tools that will help free them up to focus on other areas requiring their deep expertise.

With the digital twin approach, automated 5G network design is finally possible – and the timing couldn’t be better.

A few real-life digital twin use cases:

Use Case 1: Reducing Time Required to Update Network Parameters

One of TEOCO’s more innovative EMEA operator customers wanted to update neighbor relations across their 15,000 cell network. One of the key challenges was managing all the interdependencies between the various vendors and technologies. We decided to take an automated approach by using a digital twin model to help automatically create transcripts to write to the OSS and update the live network. This allowed us to generate faster network updates and minimize manual errors. What used to take 10 hours was reduced to just six minutes, proving that modeling with digital twins can provide significant operational improvements.

Fig. 1- Example of network design automation

Use Case 2: Reducing 5G Cell Site Integration Time by 72%

TEOCO recently helped a tier-one EMEA operator speed up their 5G rollout. Our goal for this project was to automatically generate cell site transcripts that their existing OSS systems could leverage, but this was no simple task. The key challenge was managing the complex web of interdependencies between various vendor software systems, physical geographies, device types, and a mix of network technologies.

Ultimately, our digital twin model was used to help automatically generate thousands of vendor-specific parameters for each site, reducing integration timelines by a whopping 72%. That’s a massive improvement, and this was a zero-touch process, which means it required no manual intervention. (Figure 1, above)

Use Case 3: Automating Network Coverage Map Updates

Most CSPs maintain detailed geographical maps to show regions of network coverage. These are often used on corporate websites and promotional material to communicate which areas are covered by the operator’s network. Our goal in this use case was to help our CSP customer update their maps as quickly and accurately as possible, so we created a way to automate this process using a digital twin model. When coverage-related parameter changes are made, the coverage map automatically gets updated on the operator’s website within a matter of minutes of the change being made.

Building an Accurate Digital Twin

With a digital twin, entire physical environments can be replicated- including buildings, street hardware, vegetation, and most importantly, the network itself. CSPs can also enrich their models with other sources of information, such as Performance Management (PM) statistics, geo-located measurement reports, and even crowdsourced data. Creating an accurate network model is critical as it provides CSPs with a much better understanding of network behavior and reduces the need for costly field testing.

Maintaining Your Model

One of the essential requirements when using digital twins for network planning is to keep it up to date with all the changes that occur in the live network. Accurate modeling requires regularly synchronizing the parameters between the live network (the physical network) and its digital replica. The digital twin model must also be flexible enough to update only the parameters you want without disturbing the rest. This process can be very data intensive. A large network may have over one million cell sites with one hundred parameters per cell that require updates. The key to success is not just in quickly updating vast volumes of data but also in checking for network inconsistencies – such as golden parameter deviations and business rule exceptions. Once these checks are complete, the CSP can use the digital twin as a modeling tool.

How TEOCO Can Help

TEOCO’s ASSET suite of network planning and design solutions delivers all the radio planning capabilities you need to design the best 5G network possible. 5G NR modeling with advanced propagation models, complex antennas and full multi-technology 3D coverage and capacity simulations are all supported. But ASSET isn’t just for 5G. It is one tool that does many things, including the simultaneous planning of wireless networks across GSM, UMTS, LTE, 5G and all other supported technologies.

Over the past 20 years, TEOCO has delivered the best network planning tools for every major mobile technology. We understand what it takes. ASSET’s 5G capabilities have been designed to meet your needs, especially over the next decade as 5G grows and matures; coping with the inevitable uncertainties and evolving to support new 5G technologies as they are developed.

You can learn more about ASSET’s capabilities across all technologies here, and be sure to check out the rest of TEOCO’s three-part series on 5G network design:

Part One:
Three Reasons  Why Automation is Critical for Designing 5G Networks
Author: Dr Dimitris Dernikas, Head of RAN Solutions, Support and Center of Innovation

Part Three:
Intelligent Automation for 5G Network Design
Author: Dr. Konstantinos Stavropoulos, Product Management