The State of Nokia and Mavenir, and Hot Takes from DTW Ignite 2025 (with Earl Lum)

It's been a short four months since the last episode of "From the Radio to the RIC" with Earl Lum of EJL Wireless Research and Leonard Lee of neXt Curve. The duo decided to get on the mic to talk about Nokia's Midsummer Update 2025, Mavenir's latest business and financial update, and TM Forum's DTW Ignited 2025 which took place in Copenhagen, Denmark.

In this episode of "From the Radio to the RIC", Earl and Leonard talk about the following topics and highlights:

• Nokia's new radio portfolio and the elimination of those fans!
• Nokia's SMO play and entree into autonomous networks L4 with Manta Ray
• Mavenir's recapitalization and reinvention
• TM Forum DTW Ignite 2025 - the Mecca of autonomous networks
• The SMO - the centerpiece and backbone of the autonomous network
• Modernization before monetization - the rationale for 5G SA

If you would like to set up a meeting, connect with Earl Lum at www.ejlwireless.com and Leonard Lee at www.next-curve.com. Make a request through our respective request forms. 

Please subscribe to our podcast which will be featured on the neXt Curve YouTube Channel. Check out the audio version on BuzzSprout - https://nextcurvepodcast.buzzsprout.com or find us on your favorite Podcast platform.  

Also, subscribe to the neXt Curve research portal at www.next-curve.com for the tech and industry insights that matter.

Transcript

Leonard Lee: 0:09

Hi everyone. Welcome to Next Curve's Rethink Webcast podcast and our series from the Radio to the Rick, which we haven't done in quite some time. I'm Leonard Lee, executive Analyst at Next Curve, and I'm joined by my very good radio head buddy Earl ov. EL wireless research. Hey, how's it going, man? Doing well, Leonard. Thanks for having me on. Yeah. Well, it's been oddly, a very long time since we've last chatted., I don't know why it's, I. You've been busy. Yeah. Oh, no, actually it hasn't been that long. MWC, it's just m WC is going by really quickly. But, in this episode, everyone, we're gonna be talking about and recapping ERs corporate update post recapitalization. So everyone probably heard the big news that they did get recapitalized, so congratulations to the folks at Avenir. Nokia's Midsummer launch. And finally, we will cap things off with DTW Ignite 2025, which took place in Copenhagen, Denmark, where I, suffered, Midsummer Night. Insomnia. So before we get started, please remember to like, share, react, and comment on this episode. Also subscribe here on YouTube and Buzzsprout and listen to us on your favorite podcast platform. Opinions and statements by my guests. It doesn't matter how much I like Earl, their views are their own and don't reflect those of next curve or mine. And, we are doing this to provide an open forum for discussion and debate on all things tech and radio and air interface and rf. So with that, how's it going? I think I already asked you that question.

Earl Lum: 2:10

It's going well. it's still been busy, even though wireless itself is in a lull. there's still a lot going on with regards to updates on where things are going, not only in North America, but across the entire ecosystem. And yeah, it looks like, we're seeing new products coming out and the question is, when does the chatter surrounding six G start to percolate and get louder and louder? I think we're gonna start to see that by the end of this year. But in the meantime, open Ran is still chugging along and really

Leonard Lee: 2:45

Okay. And we'll have to talk. I've been talking about open ran squid game quite a bit lately so,

Earl Lum: 2:52

yes, absolutely. Well, maybe chugging along isn't the correct term to use, but it isn't dead. No, even though a lot of people, it's morph dramatically and I think. everyone out there who thought it would go one direction has now seen it completely go in a different direction, and the leaders are still the incumbents.

Leonard Lee: 3:14

Yeah, yeah, yeah. So let's dive into Mair corporate update post Recapitalization. What were some of your takes there? I mean, I didn't attend, so I have no idea,

Earl Lum: 3:25

there was a lot of debt that was due and something had to happen in order to, fix that issue that was looming over the company since Q3 or Q4 of last year. And it wasn't a small amount of debt. I think there was over a billion dollars that was due. In terms of the creditors, everything got. Solve in a way that was happy for everyone, I believe in terms of a debt to equity, swap out or subordinate debt. at this point it looks like there's enough cash and there's not a huge looming payment. And going forward, the interest payments on the debt are going to be manageable from Avenir. and one of the stipulations Is from the companies that held the debt is you kind of not do radios going forward, or you drastically change the way you look at radios, for the company going forward and to revamp the strategy for the company into, an area that it does fairly well, which is software.

Leonard Lee: 4:28

Yeah,

Earl Lum: 4:29

And I don't think anyone ever doubted. Their capabilities on the software side. I think their hardware was actually pretty good. The problem was having someone buy it. Which was extremely difficult. So I think the core, or not, no pun intended, but the, they're gonna be focusing on the core. They'll be focusing on AI obviously,'cause everyone's talking about it. And, this hopefully will allow them to get to a level of, profitability. Quicker than the other strategy that they had, prior to this. because they're not gonna have the boat anchor of trying to manufacture the radios. With that said, I believe that they still have commitments that they're going to fulfill on whatever contracts they had. But I think they're going to be very laser focused and there's also the potential that, they're going to be doing some type of a shift towards an ODM model or, design services where obviously if you get paid NRE upfront, then that's going to lessen the amount of capital costs. Cash burn, which was the major issue for the radio group, part of Avenue. So they're not out of the radio business completely, but I think it's taken a different tack now going forward. Like you have substantial ip

Leonard Lee: 5:50

Yeah. Engineering and ip, right? Maybe like Right. Doing support and, yeah, that makes a lot of sense. I think it's important for. Everyone to recall the contributions that they have made into open, ran, just the whole ecosystem, enabling it and, their, their role as a catalyst. Now, I I think the open ran squid game. Turned out the way that you and I had prognosticate, and the game has changed quite significantly.

Earl Lum: 6:25

it goes back to January of 2023 when we first started talking about this. So it's been a, it's been a couple of years, but a lot has changed. A lot has completely altered the ecosystem. And the question going forward now is from a macro. Level who's going to be left on the radio side. and I think that's important because small cells know diss on the people that are doing the small cells for open ran, but that's really not where the money is. And the money is still on the macro side. So, I think all of the existing or future want to be macro radio suppliers out there in the open ran space. Really need to take a hard look at, again, I think it was over$500 million that got dumped into RD. Or the radio group at NY or cumulative, and that's just the drop in the bucket as to what Nokia and Erickson and Huawei, ZTE, and Samsung spent every year for their r and d to make better and better radios.

Leonard Lee: 7:30

So what else did you take away from?

Earl Lum: 7:33

Uh, again,, the core side of it in terms of where they're focused. Obviously they're still going to be doing everything in IMSI think as we go towards sa For 5G there'll be opportunities there. And on the AI side, time will tell. In terms of how competitive. And how much share they can gather, how quickly can they ramp up, their AI strategy and get their solutions out there? and so there's a window for them right now, for the next five years to achieve the strategy. And I think it's going to be a wait and see. And we'll get an update in November of this year, when they have their next analyst meeting. In Dallas. But I think again, at least the strategy now plays into the sweet spot of what they're good at, which is software. Right? Right. But you know, here's not that they weren't good at the radios,

Leonard Lee: 8:24

It is interesting that they in their open ran squid game season one venture, they went full stack. and they have full stack ip and I think that could be a really interesting portfolio of assets that they can leverage going forward. So, right.

Earl Lum: 8:47

I mean, they're still have the software for the CUDU

Leonard Lee: 8:50

Yeah.

Earl Lum: 8:51

So in addition to the core. So they'll still be participating in that side and they've done okay in that area. It was just that the. The cash burn on the radio business, which is immense. And it's just is hard to sustain, which is why prior to the advent of open ran all of the third party macro radio suppliers virtually went out of business because it was a bad business to be in because you never got the volumes.

Leonard Lee: 9:20

Yeah.

Earl Lum: 9:20

tier one incumbents got.

Leonard Lee: 9:22

Yeah. Well, no, that's great. That's a great update. So thanks for sharing that. Now what about, Nokia's Summer Midsummer Launch?

Earl Lum: 9:32

The reef shark, Silicon seems to be doing whatever it needs to be doing from Marvell and Broadcom. Mm-hmm. So, the new hardware products that they launched, which is part of the Hay Rock family mm-hmm. For 32 and 64 T mm-hmm. Higher power, wider bandwidth. Again, not a surprise.'cause that's what everyone's expecting. These are in particular though, interesting in that they're geared towards, multi operator ran sharing over in the European market. So the new radios are 400 watts, which means that each operator that would be sharing would be getting their 200 watts, which is what they wanted. And it has a operating bandwidth of up to 300 megahertz. So if one operator has more than a hundred, then the 300 will accommodate that. Plus your traditional 80 to 100 megahertz. That's typically divvied out over there in Europe. in terms of the, the C-band spectrum. the interesting thing is also is that these dual band radios are covering N 41. Which is at 2.6 gigahertz or 2.5, and also band 38, which is also in that spectrum. So we're looking at now what used to be LTE bands also being included into the radio with the C-band capability. I think that points to really leveraging the spectrum that T-Mobile has in the US Yeah. similarly over there in Europe now for all the operators that have had this TDD spectrum for a while and haven't really used it to the fullest extent that they could and to maximize and to monetize that spectrum now using massive MIMO in these radios. And I would assume that, Ericsson has similar products, if not out already or that will be coming out as well.

Leonard Lee: 11:20

Oh, cool. So I think you're turning your perspective and sentiment. I am.

Earl Lum: 11:26

and the perspective is that it looks like things are fixed on the hardware side and again, on the software. there is one other hardware product that is worth mentioning, which is the integrated passive active Antenna Plus, or IPAA plus That is using a passive antenna. with certain meta materials, technologies in it, and you're putting a massive MIMO radio behind it and you're beaming through that passive antenna. So in a lot of situations where you have to save space when you're mounting all of the antennas, this is actually able to do that with one pole mount. Instead of having to do two or instead of having to stack one unit on top of the other. So again, there are a lot of situations over in Europe where these products are being deployed. Most notably in Denmark, is one of the countries that is, starting to deploy this technology. And I think Iceland, is another country that's also deploying and would expect it, and also France. Those are the three that I'm aware of at this point in time, but I'm sure there's more to, continue as we move forward. just because rent space is money for the operator and everyone's leasing now on a tower that they no longer own. So

Leonard Lee: 12:47

this is all like brass tax type of stuff, right? The plumbing and the layer one.

Earl Lum: 12:55

No AI ran? No. Yeah, they did talk about AI for ran. Okay. Which is I think what everyone's talking about. Except for maybe Nvidia is talking about AI in ran, but also a major upgrade on the Manta Ray network management. Software that they've had in terms of it, being not only just so, yeah, but moving to an SMO platform, which is, I think, very interesting because again, the claims are that the Manta Ray, SMO capability is the only one out there that meets TM forms level four, for autonomous networks. And again, I think that's, the direction I would expect the entire industry to be moving forwards. in terms of their software approach and how they're doing the SMO. Right. And it kind of tease up, whatever you saw out at PM forum at the Yeah. You know, that you were at in Copen Hop.

Leonard Lee: 13:47

Yeah. And that was one of the big themes out of DTW. But yeah, one of the interesting things that I noticed in the Nokia approach toward SMO is that it's not open ran first. In fact, they're working on the integrations or the interfaces, that are o ran compliant, so that I thought was telling they didn't prioritize. Open ran. Right. And it makes a lot of sense because here's the thing, I wrote about this a long time ago and actually had very early conversations about the importance of something like A SMO, with Teek Amman who was at the time CTO of, Rakuten Mobile. And we had this long conversation about Brownfield integration, how. Broco 10 might be able to, figure out a way to create these inroads into brownfield environments for. their products and open ramp principles, Right. And so you need to have some degree of critical mass forming, or at least some beachhead that you set up. And an SMO would be something that would be a great tool for that, but it would have to be sold at a much higher level than the ran. Rakuten actually had an element or, a component in their overall architecture that, I think was called the Orchestrator, Rakuten Orchestrator, or RCP, you know, remember the Rakuten Communications platform? there was a layer there that was basically what we're talking about today in terms of SMO. Ericson came to market with their intelligent automation platform and that took this multi-vendor, a multi technology, approach, right? And they were taking a bit of an open ran first type of angle to it. And then I think that's shifted over time because of. How Open Ran has, played out over the last three, four years.

Earl Lum: 15:57

and I think on the Nokia Manta Ray, it's really has been originally focused on the purpose-Built

Leonard Lee: 16:03

right, purpose-built,

Earl Lum: 16:04

and then now adding the additional open ran capabilities and interfaces and such. But I think it's harder to do the SMO on a purpose built because you really have to know what everyone's, Equipment is doing, and you don't have the luxury of those open RAN interfaces.

Leonard Lee: 16:20

Well, they have the third party integrations that they mentioned, or at least I saw mentioned. And so if we think about edge Edgewise, so Qualcomm acquired cell wise, a couple of years ago and they are one of. The, leading SMO solution providers, right? They have a legacy in sa but that's one of the secret sauces that they bring to the table, or, accelerators that they bring to the table as a portfolio of these third party integrations, right? So they've worked with all kinds of different kits and so they have these adapters that you can leverage to help you accelerate any kind of integrations that you're trying to do across a, Multi-Tech or multi-vendor, portfolio.

Earl Lum: 17:06

yeah.

Leonard Lee: 17:06

Right,

Earl Lum: 17:06

and if Open Ran doesn't play out the way that the Open Right Industry had envisioned, this Brownfield is going to be critical for you to get from where we are today. To going forward to whatever that autonomous network is without the ability to have the open ran interfaces if they don't get implemented.

Leonard Lee: 17:26

Right. Yeah, exactly. And that's why, that was one of the reasons why I brought this up with Tarek long time ago, and it was that integration layer that I always thought was absolutely important because, I think one of the problems with the open round movement was it assumed that everyone, all the operators just wanted to flip. From the, purpose built to in traditional, ran to the open. Right. And that would Right. Without

Earl Lum: 17:55

understanding the barriers of what it would take for them

Leonard Lee: 17:58

transformation. Right. And so in my experience, it has always been in the IT environment. So especially if you look at. How the ran and telco infrastructure is moving, toward this more virtualized cloud, quote unquote native architecture. those types of transitions, you always have to consider the brownfield dynamic and. Very carefully plan these transformations. And so having these kind of SMO ish type of integration tools, those are the things that'll provide you that fast path to, automation and then whether or not you're open ran or whatever underneath. you can achieve probably high value levels of, autonomous capabilities. And it might not be holistic, but at least there's gonna be certain aspects that you can automate, with a certain degree of autonomy, across the portfolio versus being constrained in, in silos. So it's kind of cool to see in Nokia, evolving. Their offering. But then, okay, so now speaking about DTW, that was one of the big. takeaways that I had is SMO is really becoming that centerpiece and that first phase of, autonomous network enablement, right? It's like the backbone, right? And, and autonomous networks was absolutely the big theme. there was a huge focus on. Level four and what it takes to achieve level four. I briefly spoke with Nokia about their SMO offering, but didn't get down to the level where, and so maybe I need to have a follow up. In fact, I am gonna have a follow up with them about how they're, qualifying, that they're the first, level four.

Earl Lum: 19:54

exactly.

Leonard Lee: 19:56

SMO, but, I think a lot of folks are close. I mean, one of the things that we saw in the, catalyst there, at least I saw at the Catalyst, and if you don't know what a catalyst is, is they're these POC programs that a TM forum puts together and they're freaking awesome. I love them. Right. You learn so much. Just going to each of the catalyst and talking to. The teams that are comprised of sponsors who are usually operators and the participants who are usually vendors and technology solution providers and, GSIs or si right systems integrators. You learn a lot from talking to them and, Getting some insights into the challenges that they had in delivering, and then what worked, what didn't, that type of thing. Right. And so, autonomous networks huge. One of the things that I didn't see as prominent was all the, customer experience stuff. I mean, it was there, but it was not as a big of a topic, right? So I think this whole idea of. 5G monetization has been kind of pushed, to the wayside a bit, or at least in the back burner simply because I don't think this big bet that some of the leading O-S-S-B-S-S vendors made on generative AI is actually playing out the way that they had hoped. Right? And so this is one of the things that I had warned. Several of them about, and it turns out it is the case. I didn't see a lot of folks going around talking about how Generat, Aveda AI has transformed the customer experience in a revolutionary way. Right. And so I think there's still a lot of work to be done there. But definitely a focus on, the infrastructure.

Earl Lum: 21:44

are you seeing now the industry looking at this technology and what's been going on the AI side more inward in terms of what they can do for themselves as opposed to, again, the customer? Experience that you didn't see there that would be more outward facing for the operator and are they seeing that? Here's where the best bang for my buck is, with regards to this technology is what do I do to help my own network get better, more efficient? And to have a better TCO, right?

Leonard Lee: 22:17

Yeah. using agentic frameworks. So what I mean by agentic frameworks are these, programmatic frameworks, that allow you to create these isolated, highly controlled agents that are plugged into, let's say, a large language model or a reasoning model. leveraging these in, getting to that next level of autonomy in network management, orchestration, and, control. There is that exploration going on. That's a lot of what you saw in the catalyst. I would say that we're still a ways off. it's very exploratory at the moment, but one of the things that I think is a key. Learning that's coming out of the catalyst is that you do need to full life cycle, manage these things. And when you get into that conversation with, these folks who are doing these innovations and POCs through the Catalyst programs, you realize a lot of that thinking has not gelled yet. It's still very nascent and the industry is still exploring and discovering what. The best practices are in terms of leveraging a gentech frameworks and, generative ai, right? And so anyone who goes out there and tells you that they've solved the problem or we're at a state of revolution, I will borrow the statement that the CSO of. The Ford Company made, on stage at Cisco Live. anyone who tells you that they have the solution is a liar. So yes, And so this is the disconnect, right? That, and this, these are some of the ground truths that are surfacing through these POCs, which I think is fantastic. This is why I love these catalyst, right, is that, There we're still very much in the discovery mode and the limitations of this technology that I've written about for two years are starting to be realized. And so now this is the most fundamental learning that the industry needs to do. Once you understand what the limitations of the technology is, and granted it continues to evolve quickly. Then once you have that baseline understanding, you know what's possible, you can explore architectures and solution frameworks that. Have a higher probability of being viable. Right? But more importantly, you're going to test these, potential limitations. You're not going to discover them, through random, ideation, right? And so there's ways to. accelerate your POC and your discovery so that you can get to viable much quicker than I think, industries have been doing over the past now going on three years with generative ai.

Earl Lum: 25:16

if you looked at. What was going on and the fact that the industry is being truthful about what's going on with these catalysts and what went well, what didn't go right. Is there a sobering effect across the industry or what was the tone that you saw over there at the event? because if I go back to a decade ago, plus when. The telecom industry was looking at, network function virtualization and SDN and all of these things that the Datacom guides were trying to port over. the first year was total excitement and joy, and by year three it went to despair because the telco industry. It doesn't move in that direction and isn't going to adopt it. Or what they did was everyone adopted it vertically in their company, but there was no horizontal adoption. So it was still completely siloed in terms of the S-D-N-N-F-E capabilities within n Ericsson or nok yet, but they still didn't talk to each other and didn't have the interfaces. So it didn't play out that like the industry had hoped for because they didn't really understand the, the telecom market. Yeah. seeing something similar from what you experienced in Copenhagen or is there still this optimism that they're going to solve the issues and continue to move forward, but be more realistic about what the limitations are like you indicated?

Leonard Lee: 26:42

Well, you know, there's always, we need to on top, right? Yeah. There's always hypey themes, but when you. When you see players like Ericsson and, and props to the marketing team at Ericsson, you know, come to the conference with the mindset of no BS O-S-S-B-S-S, that's refreshing, right? Yes. And that, that's it is, I think is really what needs to happen, is like, look, try to figure out the BS ahead of time. So you didn't have to be distracted. That's another word that came up several times at the conference. We are too distracted. I said, yeah, you are. You're way distracted. You are getting excited about a technology where in order to realize these aspirational capabilities and outcomes, you need to get your stuff in order. And most telcos don't have the readiness to do all this like science fiction,, autonomous X, Y, Z, with agentic AI and all this other stuff. They don't. And then the vendors don't know how to get their clients there. It, it there that is a fact. And then again, going back to, the,, CSO of the foreign modern company, if anyone tells you that they're a liar, um, so be less distracted. Like Carl, Matt Carlson, who is the head of B-S-S-O-S-S at, Erickson, he said we need to focus on AI that is practical and works and less of the hype stuff. And that's absolutely true, right? Because there's so much work that has to be done with AI to make it work and, yield the kind of ROIs that everybody is theorizing about. And so. it, it's good to see, that there is some leadership there and it, it's not just, Ericsson. There's other companies, as well as operators, that are, Starting to, narrow the aperture of the hype discussions toward more focused things. And, again, I didn't hear monetization as much. I'm sure that there were some sessions about it, but in terms of being on the floor, talking to folks, there was much less talk about monetization, but more of modernization, which is something that I had been advocating for the industry for, a few years now, that I think is becoming that realization that the industry itself doesn't really have an appetite for hype stuff and going through the discovery process. They are with generative AI and, you know, ai, They, there, there's going to be scrutiny and prioritization toward the practical and what's actually gonna move the ball,

Earl Lum: 29:42

So what we really

Leonard Lee: 29:43

is that essentially an encouraging thing,

Earl Lum: 29:45

I think essentially what you're talking about is this m and m strategy of modernization and then monetization.

Leonard Lee: 29:53

Yeah, but you also need to be able to, enable that full stack. Like something that I had, done some work with, Amdocs on and positioning for, MWCA couple of years ago, is this whole idea of full stack modernization. So going from, the network management. That level, all the way on up to customer experience. And the idea that you do need to have that capable revenue engine, right? Or CPQ, that enables you to, be able to monetize a service that you imagine, because one of the things that folks don't. I may not realize, except for real practitioners that are deep into, the, front office operations and the financial operations of a telco. if you can't set up your systems to bill a customer. For a particular service that you've designed, you can't monetize. It becomes cost prohibitive. Your CFO is gonna basically x, you know, basically x and a it. And so the conversation around monetization requires you to, have this full stack of capabilities that are aligned with the maturity that you. Are, realizing from the network on up, So this, the stack has to be enabled, you can have a completely modernized network and your operational stack is not, you just simply can't monetize, the way that you might have imagined.

Earl Lum: 31:29

So which would come first, and if we're talking about this in terms of the modernization and this trend for a lot of the operators that started in 5G with NSA and now are moving to sa, do they need to go to SA first or do they do need to fix the BSS or S first? Or does it need to happen in parallel? Or what would the operators need to do to have both of those Yeah. Fixed in order to get to this monetization level?

Leonard Lee: 31:55

Yeah, I think so. I think they have to modernize their BSS and then OSS, really being informed by network maturity and the requirements in terms of, the required capabilities for, service delivery and service composition or design. Because, Depending on the complexity of what you want to deliver to your customer could require OSS capabilities that you might not have traditionally. these are all things that you have to, consider. As you look at your transformation roadmap, but full stack modernization, there has to be some orchestration and planning on how you do that,

Earl Lum: 32:43

I guess if the operator community's being told that to monetize, I need to move the SA so I can do my slicing and dicing and everything else, I also need to obviously have the ability to deliver that service. Go for it. Yeah, deliver for it. So

Leonard Lee: 33:01

deliver it. Yeah.

Earl Lum: 33:02

If I'm doing that,

Leonard Lee: 33:05

yeah.

Earl Lum: 33:05

I need to do both of'em simultaneously, which as the complexity now of upgrading and adding the 5G core on the SA side, in addition to doing this B-S-S-O-S-S, modernization as well. Right.

Leonard Lee: 33:19

And so, yeah, going back to the, my earlier comment about the complexity that's what network slicing and what Ericsson calls differentiated connectivity based services, that that's the complexity that gets introduced. So you do have to ask yourself. Can you bill for that stuff?

Earl Lum: 33:42

I mean, technology wise, I can enable that slice. I just do whatever on the core. Commercially or able to do that. But then how do you bill for that slice? Easily enable to that customer becomes now the gating problem.

Leonard Lee: 33:53

Yeah. Yeah. So anyways, DTW loved it. so glad that I went, didn't get a lot of sleep, but wonderful conference. I think that's all we have time for, right?

Earl Lum: 34:03

Think time's up.

Leonard Lee: 34:04

Good

Earl Lum: 34:04

chat.

Leonard Lee: 34:05

All right. Great chat. thanks everyone for hanging out as long as you have. remember to reach out to earl@www.ejwireless.com and connect with him on LinkedIn. contact him to get a hold of his reports as well as his advisory services. he is one of the leading guys in the space, so also please subscribe to the next Curve YouTube channel. And the easiest thing to do is subscribe to the next curve research portal and media center@www.next curve.com for a constant diet of the tech and industry insights that matter. And we'll see you next time.