Transcript.

Peter: Hello, Dima. Thanks a lot for joining us today. Before we dig into the questions, please shortly introduce yourself to the audience.

Dima: Hi, Peter. Thank you very much for having me here. As you've introduced, my name is Dima. I'm running the product organization for Sony Israel. Sony Semiconductor Israel, we develop semiconductor solution for the IoT connectivity. We build the modem part, the processors, the MCU, the entire SOC that can help our customers to develop their IoT solutions.

Peter: Very interesting. Thank you very much. I think that's a very exciting topic and very much in line with IoT. So let's dig right into it with the first question. LTE-M versus Cat1-based comparison. So how do they compare in terms of power efficiency, network, longevity or other matters? How do you see, how do they compare? What is the difference?

Dima: I think it's kind of a little bit similar technology. They started in the same place. At some point, Cat-M took a separate turn saying, okay, let's do something that is built specifically for the low-power LTE devices. And if I look on the LPWA or Cat-M/NB-IoT technology today, they would have much lower power consumption. I would say, in order of magnitude, lower power consumption. And therefore, longer battery life. They are built to overcome poor coverage conditions. For example, if you work with single antenna in a basement area and you never move, then we can overcome that with Cat-M technology. I think longevity is probably one of the more important questions, as well as we see now shutdown of the 2G network. 3G has gone for a while. We want to be able to guarantee the long life for the Cat-M/NB-IoT devices in the field. And that's happening.

Peter: Okay. I think you already hit the major advantage. It's longevity really and the lower power consumption, which then leads to the easier deployment.

Dima: Right. And I think maybe it's different customers asking different questions. Some devices need to live in the field maybe for — actually, some of them, only a few months. Some, maybe a year or two. For them, the longevity is not yet a question. But for a bigger infrastructure installation like utility—which could be smart meters, gas, water, electricity, smart light and then grid connectivity—the longevity becomes a super important question because they want to be in field for 10 to 15, and sometimes even 20 years. From the discussions we have with the carriers and infrastructure partners, this is probably the only technology today that can guarantee that.

Peter: Okay. Interesting. So that actually brings us to the next topic, the future proofing of IoT connectivity. We're saying that it should be future proof until 2045. Is that marketing, or is it real?

Dima: I think it's becoming a reality. We do hear this kind of commitments from some people in the industry. It's very difficult to make these commitments. But players understand that if they do want to deploy infrastructure, they should make such commitments. And we started to hear that in different regions. So it's no longer a marketing. I think it's really happening. Especially with no alternatives, if you want to onboard a utility player now as a carrier, carrier will need to offer something. They cannot guarantee such longevity on LTE. 5G has not well deployed yet. There is no 5G standalone country-wide coverage. And also, on some technical aspects, 5G may not always be the right solution. Therefore, there is no alternative for Cat-M, unless you want to keep 2G. And that's not happening.

Peter: Okay. What is the technological aspect, what is the technological difference that makes it more long-term, suitable, for IoT deployments?

Dima: I think it's the technology and the way infrastructure is being operated. I would say IoT devices does not require too many network resources. But still, carriers need to find a way to support that on their networks. 5G generated a framework called the Dynamic Spectrum Sharing. Some carriers do use it, some less. Very much depend on the constraints and the spectrum they have. But they can within their 5G networks. Some would use a separate run. Some would keep it to virtual because, really, there is not that much traffic. Some may even move to the corner. Well, Cat-M and NB-IoT technology—I didn't touch that—can also support 5G core network in the future. So if carriers will last, this is also an option. So we have a lot of tools to enable this longevity that may not exist within other technologies. I think this is one of the reasons.

Peter: Okay. In terms of the power efficiency, it's a big topic, right? You want to deploy devices that are power efficient, that can last for a long time. We have sensors out there in the wild that can be accessed. So as we just discussed, it addresses this topic already, right?

Dima: Oh, yeah, and it's already addressing this topic. We see technology that we develop and also some of our competitors do becoming better and better in that sense. I think the dream would be to have those tiny devices working on tiny batteries for years. And with the rest that we have, it almost feels like it's almost unbelievable. Think of Bluetooth devices. They work on the coils and batteries for a few years. We're almost there with several devices. Actually, for some use cases, we're already there. This is a key. I think the long battery translates in two things, or long battery life. One, it translates to your ability to really deploy devices in fields for so long. But two, it reduces the cost of installation. In many cases, you do have an electricity alternative like AC power or, in the vehicles, you may have a battery power for the vehicle. But if you install the battery with its own battery, the device uses its own battery so it's become fully autonomous, the installation itself is much more flexible and cheaper, and anybody can do.

Peter: What kind of examples would that be? I always think of weather sensors or something that's out there, difficult to reach actually. What are those most common examples where those sensors are deployed?

Dima: I think weather is one, but I think there are not that many of them yet. I think with IoT, we always also need to look into the business aspect. If you install weather sensors, who is paying for that? Who is paying for the service? How will you handle the data? If it's my home weather station, it might be a different solution. Example would start attach the grid network in my house. I already have — my water meter is connected through cellular network. And with that, they can do the reading. They don't need to send somebody to my home. And if I would have water leak in my house, I would get a text and alert that they'd go check what's happening. We can detect. So not only it becomes cheaper for the water company to send me a bill, they also get a better service which I'm very happy about. This is on the utility space. You can think of a bicycle tracker, vehicle anti-theft devices, driver monitoring. In logistic field, you want to be able to track your shipments. Really, really endless number of example.

Peter: Yeah, very, very good point actually. Those bicycles, we have them millions here in Shanghai, these rental bicycles. How do they power these things? I was wondering how does that work. But I think it's getting better even in the future. I was wondering, how do they track where they are? Because it will consume a lot of power. But apparently, this could be a solution for the future, right?

Dima: It is. Those bicycle, the cost model is solely, you know. You have the cost of the bicycle. If it's non-rechargeable, not electric bicycle, do you just leave them and somebody else takes them? There is no way to fit in into the business model somebody going to replace the battery of the bicycle. It's going to kill it. So this bicycle will have a life of six months to maybe two years, I don't know. When you install the tracking device inside, you need to assure that the battery life will be as long as the device life. And you can do that. It's amazing.

Peter: Okay. Interesting. In terms of the network, how does LTE-M compare to other cellular IoT technologies? Just concrete, for those laymen under us who don't know really the details, how does that compare? Now, we talked about power consumption. How about data transmission? What is the most tangible differences?

Dima: I think a few. I think one is, we do want to enable this. We do want to be enable the low-power feature set because smartphones are very different from the IoT devices. We do want networks to support special power, most that enables us this very long battery life. So this is one aspect. In terms of technologies, actually, unfortunately, we have a little bit too many standards. It does not help with automation. 2G is dying. 3G is already dead, I would say, in most of the world. LTE is kind of predominant in most countries but not complete coverage. We're still seeing in many places where our customers, "Hey, we do want 2G as well." LTE-M is kind of an add-on on top of the LTE network, which is kind of a new release, I would say, in the software upgrade where we enable this narrow bandwidth and ability to work with those devices.

We also see NB-IoT. For example, in China there is no LTE-M network, unfortunately. You have NB-IoT. We do see some move from NB-IoT to Cat-1 base. Not because I don't think Cat-1 base is much better. Actually, it is better. It enables higher throughputs. In China, you do have an amazing core structure for those devices. I don't always understand how it's possible, but we see that. It's magic. We already see a shift towards 5G. I think Chinese networks and for sure in Shanghai are very much advanced. The rest of the world, do you really have 5G networks in the rural area? I don't know. If you go and check with the rest of the world, 5G is deployed in the major cities. But you go out in the city, not always. And also, there is a question on the network side. Are you moving to the 5G core? Is it standalone already? So this really needs to happen for everybody to move for 5G. And as I said, even after people move to 5G, we do anticipate those LPWA network, Cat-M and NB-IoT, to coexist with 5G.

Peter: Okay. Interesting. Who's adopting those technologies now first when we look at this new technology? Which industries? Besides the consumer, obviously, there are probably industry sectors or different applications that make use of it. Who is pioneering there? Who is going ahead first? Who is leading the trends?

Dima: I would say the technology is not new. It's constantly growing, which is very nice. You see, in IoT space, it takes multiple years to adopt technology, and then you still want to keep using it for many years. So it's not new. But more and more people are onboarding using the technology. I've touched already one space, which is utility space and smart metering and grid management. Its huge vertical global adoption, everybody is going — actually, the industry moving now from the first generation, which was automatic meter reading, to meter infrastructure, where people want to. It's Gen 2.0 where people want to be able to provide more and more services, as I mentioned. A water leakage detection or if you have some wiring problem in your house, you want to be able to detect that or even balance the net. And from remote to control line, you're charging your EV or your AC because somebody wants to manage that. So for sure, infrastructure side. Other big verticals would be logistic and house as well. If I look on the market reports, I see very big portion of others. Because others does so many devices. And if I think of my office, we have a coffee machine. It could come from a third-party company. They manage it. They will never look it up to our network. They need some kind of seller connectivity. They would place the machine and be able to service it and access to it remotely. So it will use seller technology. So it's endless, endless number of examples. I would say more on a B2B side than on a B2C side. B2C side, we see a portion, I would say. Only a portion.

Peter: Okay. So it's actually opening up really a lot of new business opportunities for all kinds of different industries, right?

Dima: Absolutely. And you want to put — it can put everything today, and this is kind of an opportunity. Maybe it's not a huge market. But maybe one thing I do want to touch is, as a tracking market in pet trackers and people tracking, we do see some examples of collars on dogs that you want to be able to track your dog, his house, and if it gets lost. So we do see some kind of implications. But it's not smartphones. It's not our TVs. It's like really low throughput devices.

Peter: That dog tracking, I need to have that once it's out. It's true, yeah. So indeed, it's a lot of new opportunities coming out. If a small device can last for years on a single battery, then that actually solves a lot of problem. Because now those things are just too heavy, or you don't realize the battery is already dead. And damn it, you can't carry it with you. It's already dead. If the battery lasts several years and there's a smart warning somewhere on your mobile phone, that will be a big step forward, I think.

Dima: One is dead and that would say the convenience. Think of a dog tracker. If you have to recharge it every day like we recharge our smartphones, you're not going to use it. If you have to recharge every three months or six months, no brainer, it's not a hassle now. So why we want to make such low-power devices? It's not just less bulk; it's convenience.

(break)

If you're still listening, then you must take technology seriously. So I'd like to introduce you to our case study database. Think of it as a roadmap that can help you understand which use cases and technologies are being deployed in the world today. We have catalogued more than 9,000 case studies and are adding 500 per month with detailed tagging by industry and function. Our goal is to help you make better investment decisions that are backed by data. Check it out at the link below. And please share your thoughts. We'd love to hear how we can improve it. Thank you, and back to the show.

(interview)

Peter: So how about challenges and limitations? That's a topic I wanted to bring up as well. What are the main challenges and limitations of LTE-M?

Dima: I think one and very obvious, it's designed for the low throughput devices. Once you want to move a video or a high-definition audio, then probably you should use other technology. I would say it's even by design. The other challenges we see, and I've touched it a little bit earlier, there are too many standards. So we can work with LTE-M in Europe, in Japan, in North America, Australia, in Korea, but not in China. And if you want to do something global, then we have a dual mode solution, which is a solution but it's a different one. But I think we got it covered. Most of the countries, plus having recent addition of satellite connectivity, we can cover pretty much every place on the globe. And so it's a challenge that is being solved. The other challenge people should ask, okay, if I want to do domestic device, then no brainer. If I want to do global device, you have to check roaming agreements. Some partners would provide better than others. But we can overcome that. I think today even though we have challenges, and maybe this is the reason why we also have our jobs, I think we can overcome most of them.

Peter: Yeah, you're right, actually. Those different standards across the globe, they are the main challenge and to understand it. So how does Sony deal with this? How do you deal with that? How do you stay up to date about evolving changes globally in different regions of the world, different countries?

Dima: I think in two ways. One is following the standards, trying to promote our message also in this discussion. And you get people saying what startups are available and how you should use them. But we also invest in long term. Being a part of the 3GPP organization, Sony has a lot of — we have a huge vision of how to bring more and more IoT devices. We collaborate with many of the partners in the field to build an IoT ecosystem. Now we're already talking about how we define that within a CDG, so we can overcome many of the problems from day one. So I think it's two ways. One, we need to focus on what we can do now, how we can help customers. Number two is to bring more customers into our ecosystem. This is really by education. Not everybody in the world can do such devices. Number three, it's long term. It's what's happening within the 6G framework, we need to make sure it's properly built.

Peter: Okay. Then we have those 5G networks now everywhere developing. How does that LTE-M fit into the broader 5G ecosystem? Is it seamlessly integrating, or is there a contradiction?

Dima: There are challenges. It's on the carrier side. I don't know what to say on their behalf. They deal with this problem. They do face some challenges, how to manage the spectrum and co-existence of that. You always use same antennas, but you want to manage it efficiently in terms of spectrum management, the core network management. I do see most of them can — not just can. They do solve those problems to enable LTE-M and NB-IoT co-existence in parallel to 5G. We actually see them, LTE-M and NB-IoT were designed in a way that they are officially part of 5G even though their file error is different, physical error is a little bit different. We do see the network coexist or leave a side of 5G as long as 5G exists. This is how we say 2040, 2045. This is the expected lifetime of 5G.

Peter: Okay. So they will coexist. This is just made for different purposes then. Probably, the 5G is more for the mobile phone users, those who need to stream the videos and, like us, record podcasts.

Dima: This is how 5G is used today. I think we don't see much of that for IoT yet. There are kind of a vertical for 5G which is called RedCap and eRedCap, which will be more suitable for IoT. And even then, this is for the higher tier. For the low tier of IoT, meter and devices, very low power sensors, we don't see any alternative. Cat-M and NB-IoT will co-exist.

Peter: Okay. A bit earlier, we touched on the deployment, on the current state of deployment. Maybe could you say a bit more about what is the current state of LTE-M deployment worldwide and the regional differences? I know we talked a bit about the differences, also including China. Could you maybe talk a bit more about that part?

Dima: Yeah, I think, first of all, there is a great reference with the GSMA website with the map. Well, it's updated but there are more — actually, there are more than what is shown on that map. But maybe it's a first-stop for people who want to see the actual map. It's to go to the GSMA site and check. In terms of deployment, it's super mature in a country such as North America. All the time, I would say it has better coverage than any other network due to the coverage enhancement features. So if you go outdoors, deep indoors or outside of the neighborhood, you will have a very good coverage. We see Australia, Japan and Korea, Taiwan with perfect LTE-M coverage, Thailand. In Europe, actually, we have seen the recent developments and announcements by carriers like Vodafone. It's public. They added network in Italy. They enhanced the coverage in UK, basically bringing also very good LTE-M coverage into all the European countries now. This is for the LTE-M side.

For the NB-IoT side, I think it's very important to touch countries like China, India, Indonesia, which are predominantly NB-IoT countries. Some countries have still dual coverage. Actually, for sure, for many of the countries. Very hard to remember from the top of my head, but I at least hope I touched the most important one. Here in Israel, we have LTE-M and Cat-M coverage as well. We do the testing. We help local carrier best deploy. It's a vivid ecosystem, so it's very nice as well.

Peter: Okay. So you work with the local carriers, I guess. Right?

Dima: Absolutely. We have different networks. We want to test our devices. And what's easier than to test it at your home city and next to the office?

Peter: Yeah, makes sense.

Dima: But still, you asked earlier about challenges and I touched all the countries. All countries has different configurations, different certifications. So we have to take this into the global scale as well. It's easy to work at home, but we have teams everywhere to test and certify the devices and the chips as well.

Peter: Looking into the future, do you believe that there will be a global standard evolving out of those different standards at a certain point in time?

Dima: Well, I do have high hopes for 6G, but it's some time in the future. I think what we have now is actually pretty good. The Cat-M and NB-IoT standards are good. The ecosystem is there. So I think we're already away over there. We built infrastructure for people to really use and deploy, and I think we are quite there already.

Peter: So Sony is basically contributing with the core components for the technology. How do you see Sony Semiconductors Israel contributing to the overall evolution of LTE-M, as a final question to the topic?

Dima: Well, I think it's actually exciting because there are so many players. We work a lot with the end customers. I want to understand the use cases. We work with the module makers to understand their needs and help them to build modules. We work with carriers to understand the certificate, their requirements, specific parameters. Then we'll go kind of horizontally and work with all the infrastructure vendors. With our interoperability testing, every time somebody is doing something new, we need to test that. So we have to collaborate with those guys. And maybe one thing I did not touch, we also collaborate with companies developing the same technology. We brought the same technology integrated into our chip. It's called the iSIM. We work with our partners also to develop that and deploy and help customers to make their life easy. So I think it's actually one of the fun things about my job. I get to collaborate pretty much within the industry.

Peter: Superb. Very appreciated, Diman Feldman. Thanks very much. I hope we can talk in more detail at another point in time. Thanks very much.

Dima: Absolutely, Peter. Thank you very much for having me. I hope it was interesting.