At MWC25, VIAVI’s Chris Murphy explored how intelligent digital twins are reshaping telecom network management. Digital twins create a real-time, virtual model of a network, allowing operators to predict issues, optimize performance, and test configurations before making changes. Murphy highlighted their growing role in energy efficiency and teleoperated driving, showcasing the technology’s potential beyond traditional network management.
Looking ahead to 6G, VIAVI is leveraging its expertise in both lab and field environments to develop hybrid testbeds that combine digital twins with physical network testing. By integrating real-world network data, these models help drive research and innovation for next-generation connectivity. With digital twins reducing risk and enhancing efficiency, the future of telecom is smarter and more predictive than ever. Want to see how this technology is revolutionizing networks? Watch the full interview now.
Steve Saunders:
Welcome back to FNTV at MWC '25. I'm Steve Saunders and I'm excited to welcome Per Kangru, Director of Cloud Strategy at test and measurement and assurance company VIAVI. Welcome, Per. Per, we've been hearing a lot here about the evolution of telcos to technology companies or-
Per Kangru:
Yes.
Steve Saunders:
... techcos. What does that mean?
Per Kangru:
Great question. So fundamentally, this strategy and the transformation is really around how do you remove the friction in order to do business? And this friction is today a lot about manual processes. How do I digitize that and how do I make sure that that move into good way so you can both operate whatever infrastructure you want in a good and modern way, and then making sure that it can operate automated? So that move is really telco to techco, and then the evolution from techco over the next number of years.
Steve Saunders:
And a big part of that is obviously centering around AI and centering around the Network Operations Center, which is really the heart-
Per Kangru:
Yes.
Steve Saunders:
... of service providers, networks. Are you seeing heavy deployment of AI in the knock?
Per Kangru:
So first of all, it starts with things that are far more basic than AI, because you need to get the basics right first. You need the data. Then from the data, you need to be able to do, I mean, boring things we've done for many, many years, technology wise, which is machine learning, forecasting and so on. That then leads up to the ability to take AI decisions. So yes, those AI steps is happening right now and we are moving that into very effective ways of automation.
Steve Saunders:
What is a dark knock?
Per Kangru:
Dark knock. Well, this is really the holy grail. This is where we take the existing sites of where hundreds of people work and we just move and leave and switch off the light.
Steve Saunders:
Okay.
Per Kangru:
And then everything operates by themselves.
Steve Saunders:
And put it in space in a vacuum.
Per Kangru:
We could potentially do that, but I mean, more importantly is that, in order to do this, you really have to know what you want, and you have to understand your intent. Because operating a network for one operator to the other one in the same country is two very different things. They have different business objectives.
Steve Saunders:
Yeah.
Per Kangru:
And that's why you need to have a dark knock, which is really understanding the intent of what you want to do and moving it there. So it's not only just a matter of switching off the light and moving on and having all business processes automated, it's knowing how you want to evolve that in a good way.
Steve Saunders:
But what happens when things go wrong? I know you've got this lab as a service or test as a service concept.
Per Kangru:
Yeah.
Steve Saunders:
What does that do and how does that help with the dark knock challenges?
Per Kangru:
Yeah. So fundamentally, dark knock is normally what you would call a day two operations. Day two means that we take our live network. But before that, we obviously have something that's been engineered, normally what we call day zero. So we link day zero to day two. So if we now identify a bug and an issue in operations, can we then automatically create the test cases that allow you to fix and find these issues in that lab environment? That's where lab as a service comes in, both as a technical means of achieving this, but as well as from business model where we can then make sure that operators, even small ones, will have these capabilities rather than investing millions and millions building up their own labs and hiring tons of people who's capable of doing things.
Steve Saunders:
Or worse, deploying it on a live network without testing it first.
Per Kangru:
That's a problem you don't want to have. No. Exactly.
Steve Saunders:
Yeah. So you are essentially creating a AI safe space or a DMZ. I mean, to me, it seems similar to this concept of digital twin.
Per Kangru:
Yes. Yes.
Steve Saunders:
Is that related?
Per Kangru:
Yeah, absolutely. So digital twin is piece of technology you use to create these environments. And you can use either digital twin to harness whatever information you have and use that to then do forecasting and projections of the future, what it's going to look like, and help you build CapEx plans or OpEx plans for that matter. But you can as well in the same way use the digital twin technology single-handedly internal to achieve certain outcomes. And we do both.
Steve Saunders:
And what about AI-RAN? That's obviously a big deal at the moment.
Per Kangru:
Yes.
Steve Saunders:
Are you working on that at VIAVI?
Per Kangru:
Absolutely. And so customers of ours have used our tools to test and validate AI functions in RAN for many years. What we're now having is AI-RAN Alliance, and there's a number of, I mean, great demos at the show, which we're part of. And there have been good announcements in the last couple of days on how we support that, and that is just the start of that evolution.
Steve Saunders:
Fascinating. Per, thanks so much for coming and talking to us at FNTV today.
Per Kangru:
Thank you for having me. It's been great.