Current and Future State of Cellular and LPWAN

For those of you who don't know me, I'm Jake Harris. I'm director of Arrow Continuity Services here at Arrow Electronics and I'm thrilled to be having the opportunity to present to you our view of the current and future state of cellular and low power wide area networks, both in the Americas as well as globally.

Before I get going too far, I want to introduce you to who Arrow is, if you don't know. So we've been around for a long time, we're at a little more than 80 years. And our sales globally this year, around $27 billion. And while we have a lot of kind of statistics in terms of who we are, the way that we like to look at ourselves is that we're a technology company who helps other technology companies get to market through a combination of engineering, finance, manufacturing, and logistics services. We think that, of all the companies on the planet, we're one of the few, if not the only one, that can truly aggregate sensor through cloud solutions for our customers and deploy them on their behalf.

You know, so we worked our way through... You know, IoT has become a significant part of our business. And while it's a bit of a cliche, most of our customers are seeking to collect real-world data through sensors, send it through the cloud in some way to analyze it to drive an outcome. But we've found that the skills and attributes that organizations need to execute these types of solutions are difficult to find. You've got people that can do elements of the solution, or the design, but not the entire thing. And so that's where we feel that we come in.

And over the last couple of years, we've built what we feel is a pretty comprehensive stack of technology and attributes that can help our customers get to market just a little bit faster. And as you can see here on the slide, this really talks through all the different things that we do. So again, from engineering, finance, manufacturing, logistics, we help our customers from right at the sensory edge in terms of their designs, all the way through edge compute, continuity in billing, cloud analytics, security services, and then sustainable management afterwards. Every single step of the electronics design and supply chain Arrow can provide.

But you didn't actually come here to talk about Arrow today, you're here to learn about how we see the world changing with regard to cellular and low power wide area networks. And I'll call those LPWANs from now on to just kind of shorten it. From my perspective, there are five really key themes with regard to this space. First of all, it's rapidly evolving. You know, there's a lot that's happening in this part of the business, especially as LTE-M1, narrowband IoT are deployed, and 5G is on the horizon. We think this is gonna change significantly how our customers communicate from sensor to cloud.

Now as our customers do that, we find that the choices of technology they have to make and the suppliers that they engage with are multiplying, both in terms of both traditional participants in this business, plus new entrants, plus new technologies. It's getting harder and harder for people to make the right technology choice for them. And frankly, once they've made those choices, the deployment of those technologies and solutions are becoming more complex as customers look to deploy their products globally.

We feel that technology is absolutely accelerating in terms of you can do different things today than you could do even a year or two ago. And that's an absolute accelerant that we're seeing. And then finally, and perhaps negatively, we see that a rapidly growing market from a cellular standpoint could lead to national protectionism. When I say national, I'm not just referring to the United States, I'm referring to other countries around the world. I'll get into more of that later on but the whole point there is that the fact that LTE-M1 and narrowband IoT are such a large revenue potential for carriers and other participants, we think that there could be some protectionism down the road.

So this is why we care about cellular and low power wide area networks. And this is just really kind of some background contextual information for you. But by 2020... I guess... I get tired of hearing about the billions and trillions of devices that are gonna be connected. But by 2020 we figure that, across various different research companies, the average is anywhere from 20.8, from Gartner, to 28.1 billion IoT connections around the world.

And we assume that roughly 10%, or 9%, of those connections are going to be either LTE, narrowband IoT, or low power, wide area networks. So as you can see, the actual market opportunity and the number of devices deployed will be significant. And so again, if you look at these different research companies, anywhere from 1.87 to 2.5 billion devices that are deployed will use cellular or LPWAN as their primary communication method to the cloud. Assuming that automotive is half that, you can see roughly from 1 billion to 1.25 billion is the number of devices that will be out there with companies that are non-automotive. So again, this is an important space and it's worth paying attention to. And hopefully this gives you a sense for why we care.

Now currently this is the way we see the world today from an LTE and narrowband IoT perspective. You know obviously the U.S. and Canada have focused on LTE-M while Europe has focused on narrowband IoT. But the one thing I would say is what we've realized recently is that companies have to design with network availability in mind, and that we've had customers who have designed for LTE-M1 solutions and then trying to deploy it to a country where there is no LTE-M1 network yet, which obviously causes problems. And while that sounds simplistic and people should know better, in a lot of cases it's difficult to know because the landscape changes so frequently.

Obviously, AT&T and Verizon have led the way in the United States with regard to LTE-M1. But with T-Mobile and Sprint combining at some point in the near future, that will provide customers with one stop for both narrowband IoT as well as LTE-M1. Now, to counter that, obviously AT&T has announced that they will be releasing narrow band IoT as well at the end of 2019/2020. So really, the point here is that, in the Americas at least, you'll have access to both networks.

Now, Canada has lagged. We've seen that the LTE-M1 has been slower to deploy in Canada as Bell, Rogers, and Telus, while publicly saying they would be to market in the first half of 2018, obviously that's not the case. It's looking more like the second half of 2018. From a European perspective, Telefonica has deployed, or is deploying LTE-M1 in Europe as well as globally, and Vodafone has been predominantly deploying in narrowband IoT. And of course Asia is mainly NB-IoT as well to start. This map hopefully gives you a sense. And there's all kinds of good maps out there that you can look at that will give you a sense for the current state of deployments because really, they change daily.

Now while that's the current state, the future state is that we expect that NB-IoT and LTE-M1 will be available in all regions around the world. Because frankly there are use cases for both. It's not people...in some cases I guess you could argue that this is a VHS versus Beta kind of argument. But in reality, when you look at CAT M1, there are certain applications that really work well, especially if you're gonna roam at all. And from a narrow band IoT standpoint, if you have a static application that doesn't move and power is a major concern, then that's really the best thing for you. Now, you can use the two of them interchangeably if you so chose. But in reality, for carriers to have both available, it really kind of fits the bill for all of our customers.

Now, this is why... I want to move on to 5G. I think that people are pretty familiar and comfortable with deployment plans and use cases for LTE-M1 and narrowband IoT. But the future state of cellular is 5G. And this is really important because I think it's easy to kind of underestimate or not understand what 5G actually is. Because I know that some people think it's marketing and some people don't but there are a lot of significant benefits that are tangible that industrial customers can get from 5G.

And you can see here on the slide the five main benefits in terms of massive device connectivity, ultra-reliability, super-fast, so basically ultra-low latency. Better capacity in the networks and then finally ultra-high speeds. So while a lot of those seem like they're repetition, they're really quite different things. But the whole point is that with 5G you can do a whole lot more.

Now, in 2019, you'll see networks being deployed, and this is important. What's more important to understand is that chipsets are available in 2019. And that is... while that's good, it also doesn't necessarily help an industrial customer get to market. Because in a lot of cases, the industrial customers are not doing chip down designs because their volume doesn't allow it, or it doesn't pay dividends in the sense that you need to have millions of deployments for a chip down solution.

Which means that most customers in the past have used cellular modules or modems, and we suspect that will be the same in the future. And with chipsets being available in 2019, we expect that the following year, 2020 to 2021, you'll see 5G cellular modules be available to customers to do designs. You know, from companies like Ublox and Telit, Quectel, etc. And that really for us is the tipping point. So once modules are available, the promise of 5G becomes real to customers.

But I guess the question is, "Why is it relevant to me?" And here are the reasons why 5G is relevant. And the interesting thing for me is a couple things. Number one, 1 million connections per square kilometer. I mean, to give you some context, the average cell tower in the Americans can handle 10,000 connections. So to go up from 10,000 to 1 million is a significant improvement in performance in a network. And I think a lot of you have probably been in crowded spaces like sports stadiums or outdoor concerts or whatever, where you have a lot of people in a small area. And even though you have five bars on your cell phone, you still can't get a connection, or you can't get data to flow through. 5G solves that problem.

Other things that are really important from a 5G standpoint, they all kind of lead into the same kind of thing. So in terms of increased data rates, sub-microsecond latency rates and increased capacity, is that it allows an entire new generation of applications to reside on cellular, whereas today they have to be on other communication methods. So again, you can read through the list here but this just gives you a sense in terms of the human-machine interaction, broadband to media everywhere. Critical services and infrastructure controls. It's a great example where you've got things like an autonomous car can actually run on cellular because the latency is there to allow that application to happen.

So things that in the past didn't really... You know, weren't a great use case for cellular all of a sudden are able to. So with that comes all kinds of benefits in terms of security, really is probably the main one. Whereas cellular provides much stronger security than most WiFi applications. So all of a sudden, cellular becomes cost-effective and available for a whole new set of customers.

And the interesting thing here too is that if you think about the capital expenditures spread over more deployments, that's relevant in the sense that when you put up a tower today, you've got a certain fixed cost that goes with that. And your network has a certain fixed cost. If that's spread across millions of units as opposed to thousands of units for tower deployed, your overall fixed cost gets spread across more deployments, which reduces overall costs to customers. So the long story long, is we are really quite enthusiastic about what 5G means for the industrial IoT customer in both the Americas as well as globally.

One thing we're seeing also too is as this market continues to develop, we're seeing carriers, or mobile network operators, play a larger and larger role as enablement companies, i.e. creating products that they sell that are IoT related or providing incubation facilities or lab facilities that allow customers to go and actually create products. AT&T and Verizon have been very open with their foundry program as well as the Verizon Labs program. And more and more they're playing a larger and larger role in the ecosystem.

And what we expect to see down the road from all these carriers, and we're actually starting to see it now, is the concept of hardware subsidization, in the sense that AT&T, Verizon, and Vodafone, they don't really care about hardware. Their interest is in activations of devices on their network. And of course selling IoT products in their web stores as well but ultimately it's activations. And their view toward cell modules or cellular modems is a little bit different from what the industry has historically looked at. You know, for them it's almost a barrier to activations, and that's all they care about. And so we expect to see them subsidizing cell modules to customers at some point in the future, either directly through rebate programs or indirectly through partnerships and investments in cellular module companies.

I mentioned earlier that we expect to see some regulatory issues. And I think if you're listening and if you're watching this presentation, you know about the concept of LTE-M when roaming, permanent roaming, etc. But just as a mild refresher is that in the past all carriers basically would cooperate in the sense that if AT&T had a series of devices that were mostly deployed in the United States but also wanted to deploy some in Canada, that wouldn't be a problem for Telus, Rogers, or Bell to allow a customer to permanently roam on their network.

But as this market grows more and more in terms of customers being able to use more and more data to drive their outcomes that they're looking for, we expect to see some level of protectionism in terms of national governments being perhaps hesitant to allow foreign carriers to have permanent roaming types of deals in their countries. Now who knows how this story ends but we've seen some early indications of situations where LTE-M1 roaming agreements have taken longer than perhaps anticipated due to the fact that there are some national protectionist kind of tendencies starting to occur. We expect to see this in Asia as well as in some Eastern European countries.

Now other technologies that we're seeing that are making waves from a cellular standpoint are the concept of the embedded SIM, as well as the concept of the global SIM, and I'll talk about both of those here. Now embedded SIM is a really interesting technology. And for those of you who don't know what this is, this is basically a semiconductor. It's a silicon-based, solderable SIM that goes right on the board and allows customers to essentially seamlessly switch between carriers of their choice. Now that is a big statement that's a little harder to actually put in practice but that is the promise of the eUICC.

And so the interesting thing is in the industry today is that we're seeing, it depends who you are, obviously. This is either a threat, an opportunity, or a new business model that you can go and develop. From a tier one standpoint... now when I say tier one I'm referring to AT&T, Verizon, Vodafone, Telefonica, etc, they see eUICC as a massive threat in the sense that the traditional SIM card has been their competitive barrier. That once you put a SIM card in a machine and deploy it, that machine resides in your network and it's very difficult or cost-prohibitive for a customer to switch. Either they can deploy a new device with somebody else's SIM, or they can send a technician out to actually physically replace the SIM cards. And depending on how much money we're talking about and how much money they can save, that can be, like I said, financially prohibitive. With eUICC, that changes the dynamic in that customers, if they have the ability to change between carriers, it represents a significant threat to tier one carriers.

Now, tier two carriers see this as a massive opportunity for them to disrupt the business and perhaps participate in markets that they haven't been engaged with. And so you'll see guys like Telus and others start to deploy products where they will see mostly switch between carriers on their customers' behalf. Now they charge for the right to do that but this allows them to, using their existing relationships with tier-one carriers, to kind of intercept and create a new business model.

And then finally, and perhaps most interestingly, would be SIM manufacturers. As you know, this technology more or less impacts significantly the business model of SIM carriers like Gemalto and IDEMIA and others who have built a business based on making plastic SIM cards. If all the sudden the industry's not using plastic SIM cards anymore, that has a significant impact on their business because obviously in this case the semiconductor...the silicon-based SIM cards are made by companies like ST Micro and Infineon. Now, you still have to use the operating system from one of the SIM manufacturers but the actual hardware sale is an impact for the SIM manufacturers.

And so you're seeing guys like IDEMIA and others create new business models to attempt to provide these types of eUICC programs to their customers that allows them to manage that on their behalf. So really, depending on your perspective, eUICC represents a threat, opportunity, or a chance to create a new business model.

Now global SIMs is the other thing that perhaps you've heard about. And global SIM itself is not really a technology, it's more just a roaming agreement between carriers. So it's generally speaking LTE, not LTE-M1 or narrow band IoT. And it's situations where AT&T will negotiate roaming agreements around the world with various carriers so that you as a customer, when you're deploying your product, you'll have essentially one global price that's predictable and you know that your product will work regardless of where it gets deployed around the world. So this is almost like an insurance policy more than anything else.

From an LPWAN standpoint. I do want to touch on this as well. So LPWAN is really interesting. And I would say that the jury is still out with regard to the long-term success of this market. You know, obviously technologies like LoRa and Sigfox have significant technology benefits to customers both in terms of offering low power deployments that are relatively inexpensive but they do have some things working against them in terms of the industry's fragmented, there's no standard model. Which in a lot of cases, customers are very comfortable with, and that requires a bit of a leap of faith for customers.

You know, from our perspective, market adoption remains a concern. Now that said, you do see some people making inroads. So Senate, which is a company out of Portsmouth, New Hampshire, is doing some really interesting things from a private LoRa network deployment standpoint. Organizations like Comcast have made public announcements around deploying LoRa networks, or around the world, or sorry, around the Americas.

And so as a result, you see some investment in this area, and we think that there will be a place for LPWAN. We don't think, necessarily, that it competes with cellular. So if you look at...you know, someone like Verizon's got 2.3 million square miles of network, plus global carrier relationships for existing deployments. You know, it could be difficult for the LoRa and Sigfox competitors to be relevant in this market. But time will tell. We're absolutely keeping our eye on these players and looking to deploy solutions with them.

And of course, from a hardware standpoint, as this market continues to grow, it will attract new players. And as you can see, on the incumbents today that you know of from a cell module standpoint...and this is not an exhaustive list, just a list of Ublox, Telit, Quectel, etc. They've historically owned this market. And as this market starts to grow, you're seeing new entrants. Both things that have been...and these are all publicly announced products. So traditional semiconductor players like Nordic, who grew up in strength in Bluetooth, for example, are releasing products around cellular.

ODMs from Asia like Listram or WNC, for example, are creating and deploying products in this space. Guys like Murata continue to deploy more and more modules around various technologies. And then new players in the cell module world like Neoway, who are pure-play cell module companies, are also starting to emerge. And then finally, companies like Particle IO, who are using hardware as a way to get into the customer and then selling a cell service, all-in-one kind of package, are also starting to take advantage of the market.

So it will continue to evolve, but what we think the benefit will be for our customers, really are three things. Accelerated product development. So again, as this gets more and more aggressive, we expect to see more and more companies like Telit and Nordic and WNC and Ublox and others innovate new products more quickly that are relevant to the market. We expect at the same time to see some level of commoditization, which means that that's usually reduced pricing overall as all these organizations seek to compete with each other, which can be beneficial for customers in the short-term.

And then finally, at the end of that cycle, we expect to see some level of consolidation as these companies start to buy each other, which really follows the path of any industry, electronics or otherwise. I mean, you can look at the flower industry or the automotive industry of fragmented, becoming consolidated over time. And so we expect that this business will be no different from any other in history.

And finally, what we're seeing too is that current market players are evolving to meet these challenges. So while I mentioned Telit and Ublox, they seek to innovate to stay relevant in their business, we're seeing companies like DIGI as a great example, who are looking to do different things with their products and looking to move up the stack in terms of things that they provide to their customers. So whereas DIGI historically has been modems and phones and embedded compute products, we find now that these same organizations are looking to do different things, such as offer connectivity services, or offering smart solutions. You know, in this case smart solutions is cold chain monitoring, or call it what you will. And then looking to have more innovative embedded solutions. So in the past, hardware companies would make hardware. We find now these same types of organizations are looking to make innovative products that address market problems and solutions as opposed to fixed hardware types of solutions.

So in conclusion, we see the cell and LPWAN very much as internet of things accelerants. We know that the technology...I mean, you can really see it, will enable new applications. We think that cellular will continue to accelerate how it competes with existing wireless products like Bluetooth and WiFi, especially as security becomes more and more of a concern.

You know, we're very aware of hardware deployment with lagging network availability in the sense that... or I'm sorry, that it will lag network availability. What I mean by that is as networks become available, as I mentioned a few slides ago, the hardware will follow by one to three years, so customers have to be aware of that. And then finally, costs will continue to be commoditized. And this is relevant because as the market gets bigger, if you follow Michael Porter's five forces of competition, one thing that happens is that resales and costs get commoditized. And we expect to see the exact same thing happen here.

So thank you very much for your time today. It's been a pleasure speaking to you, and if you have any questions I'd be happy to answer them. I'm done.