One of the biggest topics at the 2019 CES is 5G — specifically, what 5G networks and devices will look like as they begin to become available throughout the world. Verizon was the first major carrier to launch a commercial 5G network and is currently serving customers in four U.S. cities with a 1Gbps wireless solution. Over the next few months, its network will both expand and add support for mobile 5G devices, competing with AT&T’s more recently launched but seemingly still nascent 5G network.
As CES 2019 kicked off, I had the opportunity to interview Verizon VP of Network Engineering Mike Haberman, who provided valuable insights into the unique way Verizon is approaching 5G, as well as how its network will evolve throughout 2019. In short, Haberman — an engineer responsible for the network buildout — says that Verizon is taking the idea of “true 5G” seriously, implementing the 5G standard’s highest-performance elements to provide a different and better customer experience than its competitors.
Here’s a lightly edited transcript of our discussion.
VentureBeat: How will the transition work from Verizon’s current 5G TF home broadband network to its 2019 5G NR mobile network?
Mike Haberman: We launched what we call 5G Home in 2018, and that’s a real system with real customers. That’s real 5G, using TF standard, technology form standard. Obviously, what happened is that TF was a bunch of the bigger infrastructure vendors and us leading the way, and we drove the industry to get 5G out, and what ended up happening was that the 5G NR standard was accelerated. Which was great from our standpoint. From a network perspective, all the ingredients are the same.
It’s merely just upgrading that last leg, which will happen, and we’ll do that when it’s ready at no cost. It’s just a modem upgrade at the house, and it’s designed to be upgraded, and some of the equipment can be upgraded with software. It’s all millimeter wave (mmWave), so it has the same radio. We’re not using a different radio; it has the transmitter, same receiver.
So the plan is when mobile is available, we will be very aggressive in deploying it. No change in stance there. From my perspective, I’m building in all sorts of places across the country, to really get ready for that when the September (2018) release standard hardware is available, which is happening right now. What’s going to happen is that when there’s (consumer premises equipment) CPE available on 5G NR that can do 5G Home, then we’ll introduce NR gear that can do 5G Home. There will be an upgrade path that will be fairly straightforward. We’re moving forward quickly.
At the end of the day, there’s going to be one (Verizon) network for 5G home and mobile. There’s not going to be two different networks. That’s not the plan. We’ll have one that can be scaled, with different attributes for different kinds of services, that’s the way 5G is designed.
VentureBeat: How many customers are there right now on 5G Home broadband?
Haberman: I’d love to tell you that, but I can’t. We have an earnings release coming up on the third or fourth week of January.
VentureBeat: Right now, Verizon’s 5G network is strictly using mmWave frequencies. AT&T and T-Mobile have espoused the value of multi-frequency solutions in building their nationwide networks. Is the rest of the Verizon network buildout going to be mmWave, or what frequencies beyond mmWave will provide the ability to do your nationwide buildout?
Haberman: The question is, what’s 5G, when it comes down to it? You can take 5G and try to put it in 600 MHz (spectrum, like T-Mobile), a lower-band frequency, but technology-wise that doesn’t do anything for you. You’d wind up with a 5G NR carrier that would basically have worse throughput than LTE-Advanced, because you wouldn’t be able to aggregate bandwidth together. If you want to create a marketing message that doesn’t benefit the customer, you do that, but that makes no sense to us. So the question is, what are you trying to do with 5G?
There are really two underlying principles with the way 5G was designed. One is massive bandwidth, so you need to have something on the order of a 100 MHz carrier for NR to start to be meaningful. Otherwise you’re just talking about LTE-Advanced. And mmWave is obviously a place where you can get a lot of bandwidth.
The second thing is antenna technology. When you look at mmWave, one of its negatives is propagation, that (its signal) doesn’t propagate as far. But one of its positives that is going to become very apparent is that the antenna technology you can bring to bear with mmWave is immense. You can do a 64- or 128-antenna array where the beam of the antenna used to rely upon three antennas 120 degrees apart.
With 5G, you can have 128 antennas in the same space, and you’re going to have a beam that points directly at you to reduce interference and make sure you can communicate clearly with the cell site. So that will help as to how far you can be away, because we can be very spot on accurate with the beam coming at you. You can really only do this with mmWave, because at higher frequencies, the array of antennas would be massive.
A true 5G network needs to have massive bandwidth and you have to be able to use this antenna technology. There are shades of gray, like you can talk about mid-band and what have you, but the truest spec was meant to be mmWave, to take advantage of all the antenna technology. From our standpoint, what is worth looking at is the density of our sites in these bigger cities. You’d be amazed at how dense we are right now.
If you have the densities required, and you bring in the antenna technology, and the bandwidth, it’s like, “wow, that is a different experience.” We’re interested in providing a different experience, not just more of LTE-Advanced, that’s not what we have in mind. It’s to provide a better experience, to change the dynamic in the industry. That’s why we’re looking at mmWave, and we’re going down that path, because that’s truly what 5G is.
VentureBeat: AT&T and T-Mobile have suggested that they can’t blanket enough territory with mmWave hardware to go nationwide. Is the Verizon theory going forward that you’re going to move slowly, but every city that gets Verizon 5G will get mmWave? Or is there going to be a fallback, apart from LTE-Advanced?
Haberman: People have talked about sub-6 GHz. You’re going to look for opportunities like that as (5G evolves). As you get more of the handsets that have (5G) NR capability, then it makes sense to migrate some of the other bandwidth you have and then aggregate them together. The problem you have (now) is that you can’t aggregate (existing spectrum) together because you don’t have enough (capable) handsets. But as you get more penetration of NR-capable handsets, all of the sudden you can do things that are more meaningful with the frequencies.
When you roll out a network, which god knows I’ve done this three times already, you can’t just do everything at once. That’s why this first step is truly the way to go, and then you’ll add elements on as you move along. The statements I hear from other carriers are just them trying to realize the limitations of what they have, and trying to create a story that sort of makes sense.
VentureBeat: So we’re not going to see from Verizon in the immediate future a variety of 5G flavors like “5G+” or “5G Enhanced?” It’s all going to be branded under the same name, you either have 5G or don’t?
Haberman: I can’t comment on that at this point. Clearly we’re going to come out and clearly say what 5G is, though. It’s going to be what I just explained to you. Any changes that are done, we’ll clearly communicate it. From our standpoint, we’re defining what 5G is, and it’s what I just told you it was. Because there could be a lot of confusion there.
We saw what AT&T did with “5G E” and you saw the reaction to that — it got the reaction we’d expect, quite frankly — the same thing that happened when AT&T upgraded from 3G to “4G” overnight, they did the same exact thing, and it didn’t help them. It begs the question, “what’s different?” They’re a laggard in terms of performance already, and they just change the name of it and say it’s better, but it’s still worse than our LTE-Advanced technology.
That’s going to come out in the wash. People are going to compare 5G Evolution to LTE-Advanced; I bet you some third parties do it and it’s going to be rather embarrassing when our LTE-Advanced network is faster than “5G E.” How do you explain that?
VentureBeat: Regarding the state of your network, where will things be by mid-2019?
Haberman: I’d love to tell you. But all I can tell you is that we’re very aggressive, but we don’t want to tip off our competitors yet as to where we’re going. The ingredients to make 5G work, there are really four ingredients. The first one is densification of the network, all the 4G densification, which has been going on for years. The second one is fiber, specifically dark fiber, because if you think about the bandwidth demands… we knew they’d get higher and higher.
And then, we’ve been buying spectrum in millimeter wave, we bought it from Straight Path and XO Communications. Then the fourth one, which has to deal with reducing latency, is that we’ve been virtualizing our core. It allows you to move the components of the core out to the edge of the network, so you can get pretty close to the cell with core network elements, and improve latency. Devices will become a lot more snappy, and it also enables applications like AR/VR.
We’ve been working on these four underpinning enablers for a number of years to make it happen, which is why we’ll going to be in a position to be very aggressive with 5G NR millimeter wave.
VentureBeat: So you’ve built out this dark fiber network and anticipate tremendous bandwidth utilization. There have been forecasts that unlimited data was going to be basically necessary to facilitate 5G. You’ve already offered unlimited data for home broadband; is that going to be the same thing with mobile 5G?
Haberman: We have not announced the plans for mobility and have not made them public yet, so no comment on the pricing plan. That will be something that you’ll see shortly.