 So, our next panel, third panel for the day, is going to be moderated by Dean Meng Chang. It's hard to introduce a person like Dean Meng. He has, wears multiple hats here at Purdue. He's a Senior Vice President, Executive Vice President for Strategic Initiatives. I have the good fortune of working with him a lot on that. He's our Dean of Engineering. I have a wonderful opportunity to work with him on that as well. And he's also a professor, a chair professor in the College of Engineering with a specific focus on a lot of the topics that are near and dear to what we're trying to do here in Discovery Park with regards to network, advanced network development and deployment. Very few people also appreciate the fact and recognize the fact that for the last year of the Trump administration, President Daniels sort of lent Dean Chang to Secretary of State Mike Pompeo as his Chief Technology Advisor. And it was in that role that Meng came back to Purdue, the wiser, the more connected to all of the problems and challenges in his portfolio grew accordingly because of that important role that he served the United States in that year away. So my pleasure to introduce one of my favorite colleagues here at Purdue, Dean Meng Chang. Well this one's going to test the connectivity here in the Convergence Center here. I hope you can all see from the screen here, we've got four outstanding colleagues. And to those who are joining us here in person and online in this hybrid format with streaming and also recording, well, first of all, thank you, David, very much to introduce me to moderate the concluding panel session today. And I assure you that after spending one year in Washington, D.C., I came back one year older, not sure about wiser at all, but glad to have the chance to moderate this panel. So coordinate wise, well, right now we are in Convergence Center to those who are online in the Discovery Park District of Purdue University. This is a connected community, lab to life, as we just heard in the last panel. So time-wise, we had the wonderful outstanding demos, outdoors, indoor Jungdans, we've had poster session, innovation pitch session, we've got opening by President Mitch Daniels whose vision started the whole Discovery Park District. And then we've got Food Truck, I heard, the frozen yogurt and custard were all sold out and the rain still hasn't arrived. I don't know what's wrong with the weather forecast. We need big data to be better in the weather forecast. And then we've had a mayor's panel, we've had a lab to life panel and third and concluding panel, a building upon 5G is talking about what about 6G coming up. So that is the topic and I've had the honor to work with these four outstanding colleagues. Let me introduce them to you before we get into the conversation here. So clock-wise on screen we have Jung Saundersstrom who is the head of technology for Silicon Valley Ericsson. Thank you, Jung. We have John Smee who is Senior Vice President for Engineering at Qualcomm. Thank you, John. We have John Rose who is the Chief Technology Officer of Dell. Thank you, John. And we have Asha Kedi who is Vice President for Engineering at Intel. Thank you, Asha. Well, I am going to sit down in this very comfortable chair, although I may not be able to see you. Ah, I can look at this screen here. And you see some empty chairs here, pretend they're here. They will get to a holograph and you'll see them and you can smell them and all that. But for now, you can have to look at the screen in this two-dimensional rendering. So 6G, maybe I'll just float the first question here in no particular order. But maybe we start with clock-wise and then we'll go back to counter-clock-wise. Next question. What was through deploying 5G? If you're looking at here in the Middle America, in the Midwest, in Indiana, or in the city of West Lafayette Discovery Park District, what is the things that 5G is not delivering but we need 6G maybe by the end of this decade? Maybe we'll start with John and unmute. Thank you, Mike. Thank you. All right. Thanks for having me on this panel. I really appreciate to be here and to talk about this journey. I think Indiana is in a good spot for taking use of this journey we're now into from 5G to 6G and even beyond. And maybe if I start sort of in the digital divide and the rural broadband side, we are with wireless technologies enabling now to bridge that divide much faster. Always when we look into mobile or the broadband divide, mobile broadband divide versus fixed mobile broadband divide, the mobile broadband is always ahead in terms of some 10% of closing that divide. So now going into deploying 5G for fixed wireless and later 6G, this is a faster and cheaper journey, at least in many, many communities to leverage that wireless aspect. On the more advanced use cases for like manufacturing or for healthcare or for agriculture where you really need the super low latency and the super high bandwidth, we are seeing quite a bit of progress already today with the 5G system we are building. But of course with the even more immersive and even more low latency and the even more high performing 6G network, there will be additional. So I think it's not a stepwise function here, it's going to be a gradual. I think it's very important to develop the use cases during the time already now when the 5G is coming out. So we can start to learn how much we can push the use case given the platform that is now rolled out and then upgrade, update and innovate further on the 6G. But of course you think about going from millisecond latency even further down, right? So we look into manufacturing, the manufacturing connections with IoT in general will double over the next few years and probably over the tenfold over the decade. All of these IT connections cannot be wirelined, most of them will be wireless and the quickest growing segment there is cellular. So there I think is a huge potential in the manufacturing section. I'll stop there, I'll come back with more I guess later on. Thank you, Janna. So you're saying that in 6G the latency will be so much smaller but I assume that it will only be useful for machine-to-machine or mostly useful for machine-to-machine communication in the like of manufacturing or factory floor or warehouse because the human reaction time is not going to benefit much further from further reduce the latency. Absolutely, I mean for example we are doing a lot on VR and XR here in the lab right now and the typical turnaround time for making people not busy is like 10, 20 millisecond and that we can do with 5G today. But if you're taking industrial protocol that today's goes on a cable in a factory, they're maybe down to 2 milliseconds or 4 milliseconds and that will just enable that to convert the current systems into wireless but then of course if you re-enact it further with AI and data collection there might be unknown use cases that would require 1 millisecond or 100 microseconds or whatnot. So that is you're right, human brain will not need the microseconds but the machines will. Thank you, John. John, please. Yeah, I think what's interesting is a good question because you look at we're kind of at the halfway point of 5G if you're thinking about standards, we're at the beginning of 5G if you're thinking about deployment, 6G is not going to show up until 2030. So we have another 8 years of 5G and in particular we just finished releases 15, 16, 17, about 5 years of standardization and so there's a lot of content already in 5G and what we're really focusing on in the next 5 years is to make sure that we're basically making the right investments to deliver on some of these use cases that can be more transformative and so we can look at connected factories are enabled already in principle with 5G but there's a lot of ecosystem work to make that happen. There's a lot of trials, there's a lot of kind of understanding even what's the business model because we're not just selling connectivity in megabits per month or megabytes per month or terabytes per month. We're really selling in some sense a value where it could be more worker productivity. It could be better medical outcomes. It could be improved classroom learning or learning at home or on the go and so I think one of the interesting things then is the 5G itself, there's that mobile broadband which really also includes fixed wireless. I would say that's one, particularly when we look at, you're discussing like Indiana and some of these rural opportunities for that last mile or last 200 meters. Then all of a sudden the data rates we're doing on wireless are very similar to what many people are getting on wired. So I think what's interesting is that paradigm of what's happening on wireless and what's happening on wired, well that's always changing. I think one of the things we've seen as we move through 4G into 5G and as we're continuing to invest in 5G, there's the Moore's law benefits to how much silicon processing we can do on device, the fact that we can have machine learning, algorithms, neural networks on device. We also can obviously have those at the edge of the network and in the data center but this movement of processing towards the edge I think is also very interesting because it opens up new use cases and new opportunities where then wireless is going into some of these new requirements and maybe one last comment is I do agree also with your beyond human requirements for latency and reliability and even time synchronization. Also when we look longer term, you can imagine this more immersive world, a digital twin of the physical world and all of a sudden at that point we're designing for a 2030 time frame of really what is the future of enterprise all the way into that future. So I think that kind of brings in this use case aspect and then these technology capabilities aspect and then kind of how we look at the big 5G evolution versus the longer term bucket of 6G. Thank you, John. You and I can talk about edge compute and also with other panelists all day long but you also highlight an interesting point that the data plan might also look different. Right now we either buy a bucket of data or it's unlimited data but the charging mechanism, the monetization happens by counting bytes. How many bytes have you consumed whereas in the future in 6G maybe the data pricing, the economics will also be different. Well, John, just like the mayor's panel earlier today where we've got two panelists sharing the same first name. So we've got John from Qualcomm and John from Dell. John, what do you think? Sure. Thanks for having me. First I'll echo the other John's comment about 5G is fantastic for rural America. I mean, the fact that we're equalizing fixed and wireless performance levels just does a fantastic job of giving us the ability without a lot of hard infrastructure of just reaching more people. So I'm very bullish on 5G as it exists today. However, your question was well, what isn't it doing that brings us to 6G that will help the less dense environments or rural America or the economies that exist outside of large cities and three come to mind for me. The first obviously is in the 5G ecosystem today and cellular in general, the economics of connecting devices is still too high. Today the cost to put a 5G radio into something means that even in the machine world the only thing we put it in are $30,000 robots or $1,000 PCs. We aren't quite there of being able to put it into a $10 sensorized entity. There's work going on in the current 3GPP work and you'll see progress on that journey, but a North Star for 6G is look, we really want to connect everything. And in order to do that, we're going to have to think about low power, low cost. We have a project going on right now that says, what would it take to bring the effective cost of embedding 5G connectivity at some point or wireless connectivity into everything that comes out of a Dell factory, anything that we sell? Non-trivial tasks because it's not just the cost of the radio and it's the cost of provisioning and the cost of many other dimensions of it. The good news is I think as we move into 6G, we're all taking that very seriously and we will find novel ways to basically bring this connectivity to far more than just the let's call it the big things. It will become connectivity for the little things, which actually turns out if you want to have the trillion node internet, that's what you have to do. So that obviously will have huge implications in rural America where we have very large industries like agriculture and mining and forestry that benefit tremendously. If we can digitize them, what we're digitizing is against something that has effectively very low per unit cost of the thing that we need to digitize so we can't add a $1,000 sensor to it. So I think that that'll be a particularly interesting thing that will happen and improve our overall outcomes. The second is, I think it was mentioned earlier, the current networking paradigm for a dense urban environment might be fiber everywhere. You can get away with that. You can't do that in rural America. You need a collection of networks working together. And today in the 5G era, we actually have obviously 5G wireless and fixed wireless, but we also have Wi-Fi networks and we have satellite technology and Leo constellations. And there's a lot of technologies out there that are very useful, but they're all fragmented. And one of the design principles as we move to 6G is, can we become the network of networks? Can you have a control plane across all of this? And the benefit to rural America says, instead of you having to choose between satellite and terrestrial wireless or other technologies, you can start to think about them as a system. And we do believe the best probability of creating that network of networks is basically building off of the 5G architectures. They're the biggest networks in the world. They're the most diverse ones. As many of you know, the current 5G networks are very different than the 4G networks in one big dimension. And that is that we have low, mid, and high-band spectrum, which have very different radio interfaces, very different architectures, all working together as a relatively coherent system and actually interacting with the same end users as a fairly seamless experience. So that'd be number two on the list. And obviously, that will help us get coverage. It will help us apply connectivity to many more spaces without having to be ambidextrous in terms of thinking about the technologies at the user perspective. And then the third is really around functionality. And I think it was mentioned earlier, the 5G is great, but it's not quite at the latency levels that we need to solve some of the more advanced problems around AI and distributed processing. But it'll get there as we move into 6G. We definitely know that we're going to need significant reduction in latency, a significant improvement in performance, and clearly an increased reliability to bring more and more of the sensorized industrial manufacturing world onto these infrastructures. Now, again, during the 5G evolution, as John said, we're not done. We're halfway through the standardization and we have a lot more in front of us to roll out. It will get better. But if we use 6G right, we will have hit a point where we feel like we have significantly improved latency performance levels that will be exceeding, hopefully, the capacity of terrestrial infrastructure in many cases, which will open up, again, for rural America, lots of interesting disruptive technologies. One, you mentioned earlier, if any of you are familiar with holographic technology, it happens to be a very, very bandwidth-intensive activity. Putting a holograph out into an environment can be tens of gigabits per second, potentially, to do it right. Now, do you need a holograph today? Probably not. But imagine 10 years from now, your telehealth experience not being a clunky experience over a phone or a web browser, but a holograph economically delivered to a rural healthcare setting so that an elderly person can interact with what appears to be a human being, as opposed to some technology interface that's alien to them. There are huge opportunities, but they're contingent on low latency, high performance, high reliability. All of those things are coming and they'll be the fundamental underpinning of increased digitization, but being able to do it over distance and to be able to do it wherever we need it, which is much more of a profound issue in rural America than it is in kind of dense urban environment. So definitely cost, end-end network of networks and this continued path towards functionality, which allows us to not have the haves and have nots on the premium experiences that, quite frankly, are the underpinnings of most digital transformation. Thank you, John, so much. I want to come back to two of the themes you just mentioned. Later, one is heterogeneity of network networks as a feature in 6G coming up and the other is holographic or other applications. Imagine a virtual lab where the professors are holographic, right? Or maybe the students are. Well, but first before we move on, you mentioned the first of all the question of cost, right? And we put some numbers to say, well, maybe by 2030, the latency will get down to around one millisecond, where 5G is not there today yet. What about the cost? What kind of a factor of reduction of the cost of connectivity do you foresee to be achievable by 6G? Well, I think we have a lot of challenges there, because there are industry structural issues. For instance, today, a little known secret is that, hey, this is a very IPR-rich environment and the cost of just licensing the intellectual property to do wireless is quite high. It doesn't mean it isn't valuable, it doesn't mean it isn't important, but we need to find different IPR regimes. And it's not to devalue IPRs to just make sure that everybody makes their money back on their R&D investments and their intellectual property. But that is a problem we have to go solve, because if we don't, the base cost will start at $15 or $20 a unit, no matter what we do. And so again, not a good or bad thing, just a necessary thing we have to work through and still effectively reward people for inventing things and owning intellectual property. If we can't solve that problem or come up with a different approach to it that's equitable and fair and empowering, then we may not get to the level that we need to be, because candidly, we have a strong opinion. We believe in Metcalfe's law, we believe that the value of the network is equal to the square of the connected users. The more users you can put on the network, the more value it will create. The problem is today, the effective cost, and it's not just the cost of the electronics, it's cost of provisioning, it's many other dimensions of it, is still too high to bring it into the really low cost consumer or industrial footprint that we're going to need to get. So that's a burden for all of us and we're all working on it. We are finding novel ways to bring the cost of electronics down. We're doing interesting things as an industry to say maybe we don't change the cost of the componentry, we change the cost to the consumer, which you mentioned earlier. For instance, economic models in which the consumer may not actually be paying for the 5G connectivity, the services that run on top of it may be subsidizing it, so that again, we can remove that cost equation. But if you thought that 6G was purely a technology exercise, it isn't because one of the biggest challenges we have is how do we get it everywhere? And in order to do that, there is fundamental investment in technology that's being created and needs to be valued. And in order to actually hit the economic points, we have to find novel and creative ways to basically distribute that cost in ways that allow us to enter new price points. So for any of you who are economics focused or economists, you're going to be just as busy as us technologists as we move into the 6G world because ultimately, if we can't change the cost curve and expand the economic footprint, it's going to be very difficult to build the multi-trillion node internet, as many of those devices sit below the floor that we can currently serve with traditional telecom. Very interesting points on economics. Thank you, John. And Asha, please. Oh, thanks for having me here. And I agree with all my fellow panelists, especially the shout-out to Morsla. And so with that, I'll take a slightly different approach on Indiana. I think the difference in by 2030 to Indiana would be, first of all, we'll have it fully connected. 5G did have to do this transition from phones and then focused on the rest of the aspects in 6G. It's really going to be different. And we haven't talked a little bit like on the IoT aspects like for John Doe or Jane Doe and the con or cattle and how do you track and if it goes off and the fines and also it can optimize. But also, given the manufacturing, the factories and solar, the wind energy, which is very big, right? If you look at the transformation of those industries with networks that once are connected, the real vision things in Indiana will be the one, the people who are 10-year-old today are 10 to 15-year-old today, which will come up with innovations that I can't dream of. That will help in the things that Indiana has to do. So I'm taking a more aspirational route. And to me, with 6G combined with what we can enable with compute, if we could just computing AI and all of it, they will help create the future of innovation and economy. That's an excellent point to thank you, Asha. Let me follow up on that. And this time, the second question will go the other way around. Starting with you, Asha, picking back on what you just said in terms of the talent, in terms of the creativity of building applications, whether it's for humans or for machines, off of the 6G's capabilities and functions and latency. If you have to name one particular application, what would that be? Now, while you think about that, to those who have been following us throughout the day, you also remember what the mayors mentioned two hours ago. And that is true. Yes, we have outstanding agricultural land and farmers and a farm economy here in Indiana and Midwest. But we're also building out to densify the connectivity, such as this Discovery Park district itself, right? It is a densified connected community of work, learn, live, play together. So in your example, it might be related to the agriculture and the rural side. It might be related to the smart cities, just the smaller smart cities in middle America. But if you have to name one application, you say, if I have that in year 2030 based on 6G, I would love 6G. What would that be? Please. That's a hard one, but I'll take a more abstract approach because I'm a huge thing of sustainability. So I think the one thing that I would like to see is, we say the word digital twin, but it's a stereotype to it, but that's the application that I would go for, my own Jarvis, because we will have data, compute, and learning. So I don't have to waste water, or whether if I'm manufacturing, I can like build electric cars or autonomous cars because even the car manufacturing process is going. So basically, essentially having Jarvis, whether it's for my factory, or whether it's for my agriculture, or maybe it's for my software startup, I don't know, right? But to have that level of intelligence at my fingertips with things that I can't do would be the one application that I would see, which we can have because we have ubiquitous computing, right? Like at a scale that we've never had before. So that would be it. Thank you. And John from Dell, what do you think? Yeah, I always get nervous with these questions. When I was at the CTO of Nortel back in I think 2007, I was at Mobile World Congress on the CTO panel and they asked us what would be the killer app for 4G? For context, 4G went live in about 2011. And that's a really hard question to answer. I actually think I got it right because what I said was, I don't know what the killer app will be, but I do know that we're going from a technology that's basically TDM over the air to broadband over the air. And when that happens, every user and application that is currently constrained to the wireline world will suddenly have the opportunity to go mobile. And that's kind of what happened. We didn't know exactly what the user experience would be, but we suddenly put broadband over the air, which we'd never done that before. And so if I use that same logic for 6G and say, okay, well, I'm not going to tell you which application, but what's the difference between 4G and 5G? Or in this case, well, the 4G ecosystem is broadband over the air. And primarily what it did is it allowed us to take the human entertainment and productivity connectivity experience and make that mobile. That's kind of the value. You got to do YouTube, you got email, you could stay connected when you were working, you could do a Zoom call. And that's kind of the sum total of what we really got out of it. There were other use cases, but the real economic value was that. If you go to 5G, it's different. 5G allows us to, I would describe it as take the deep enterprise and make it mobile. The things that we haven't done yet. We haven't taken the factories. We haven't taken the transportation networks. We haven't taken healthcare. Those things have not really mobilized yet. And they're now in a position as we look at what 5G and ultimately 6G can do, that the underlying network is now optimized for that. I mean, we've been very bullish that I think I was an op-ed in the Wall Street Journal about this since people are complaining that 5G wasn't moving fast enough. It wasn't having a big enough impact. And I had to remind people, the purpose of 5G is not to make YouTube faster. You'll get that. The real purpose of 5G is to actually bring mobility to everything else, which is every device, the enterprise, the industrial use cases. And most consumers won't see that directly. They'll see it indirectly by better transportation networks, by better healthcare. And so I think that's kind of the killer app that we're looking at. It's not necessarily any one. It's just that there's this class of applications that have been excluded from the mobility revolution. And they're primarily living in deep enterprise use cases. And they are now all becoming unlocked for the first time in our history. And that is incredibly disruptive and incredibly powerful. That's why most of us are pretty bullish on the current 5G, 6G roadmap. Thank you, John. Some friends asked me, so, Monk, this 5G thing, I've been to watch Netflix. It's not going faster. I said, well, you should hit fast forward. And that will go faster. But well, John, what do you think? Price relative to consumer and flat screen, right? So the kind of, as we looked at the 4G to 5G, this perspective of being able to do more with your flat screen and then a partition to compute, as in what's happening on the device, when are you accessing a website where the information is largely in the cloud and you're just moving the application aspects around? Versus when we look all the way to 6G, I think the KPI that's going to change the most is the ability to distribute compute in new ways at the edge. And so in particular, we always have to pay attention to power consumption, size, weight, power. So when we talk of transforming how productive people are, a hospital of the future, and you look at all the sensors, the long-term data, the short-term data, the combining the cloud knowledge with a bunch of sensors on a patient. So all of a sudden you have different time frames of short, intermediate, and long. And then you also have different latencies of how quickly you need to fuse information. And I would say the promise of the 2030 era 6G communications platform is that the reason we're going to be communicating is moving beyond the human requirements, going back to your first point, that the human requirements are, we are relatively limited in terms of our eyesight and our hearing and our ability to engage with multimedia. But as we move in towards a more immersive experience, then that really challenges the question of where compute is happening. And also it challenges power consumption that even for augmented virtual reality today, there's many scenarios where you're going to not process it on the glasses, but in a smartphone. Or you could be moving that processing towards an access point. Or you could be processing in your laptop. So this ability to move processing around and what sort of data rate requirements that that brings with it, I think that's how I'm looking at what makes 6G different than 5G. It's about, you know, these different human productivity parts. And then that machine connectivity. Thank you, John. And Jan, back to you. What application? In case you're wondering, that's one of the old, you know, golden black balloons that just popped. Very excited by the topic we're engaging the conversation with. So what do you think, what would be the single application that would excite you most in 2030? A great question, and yeah, I think I'll answer in two, in two-fold. One is sort of what I, specific question, what I would like to see in 6G as a single application. And the other one is maybe what I believe actually will happen. What I think would be a best or the key thing for the society is to bring a bridge, this digital divide, immersive healthcare for everybody, education that is equal across state and countries. 6G has the capability to really bridge society, which I think is super important in the time we're living now. Now, that is maybe a personal and partly also sort of corporate view on what we want. In the same time, of course, the learnings from this industry is that the use cases that actually happens are the ones where there is someone that pulls volumes, pull volumes of chipsets, pull volumes of, I mean, John, we're into this, pull volumes of devices, so driving the whole machinery with them. And you're struggling a little bit with 5G, not seeing the industrial devices and applications and really kicking off. So what are the volume drivers for 6G? I think the answer, at least looking into where people are investing right now, comes into this AR, VR metaverse thing. And this probably starts in the consumer segments, but hopefully such a strong movement that it will pull with it immersive use cases also for all the other industry segments. So that's more of my looking into it. But if I were to choose, I would cross the bridge, the digital divide. Wonderful. By the way, since this is streamed and recorded, it will be on YouTube for eternity. So soon it will be 2030. We'll play this back and see where we are. Now, another feature of Purdue's new campus here is lab to life, meaning that any cutting-edge technologies can be deployed here first because the bureaucratic red tape paperwork is a lot less than the big cities. Now, back to the heterogeneous network component you just mentioned. Well, what about the private networks? There is the public mobile networks that most consumers recognize, but then there's also the private networks, especially in the 5G to a 6G world. For example, a private network managing airport and managing the public safety side of the city or a private network within a manufacturing facility. And we have some innovative industry 4.0 testbed here on this part of campus. So what do you think about the promise and the false of private networks in a lab to life environment? And maybe this time we'll start with John Smith and then go to John and Asha and then back to Yang. I've had networks specifically. And one of the interesting things then is the ability of the networks and the devices to be more adaptive and be more intelligent. So we've always had sophisticated algorithms for things like scheduling and using our scarce resources in wireless. I think what's interesting is we can become much more data driven. So if we look at a private network, the reason it can be more transformative in your great example of that lab to commercial environment where you're experimenting stuff. You're doing real-time learning and you're also doing real-time deployment. Well, the system itself, in some sense, is living and breathing. And it's going to have more data. There's going to be data collected at the device. There's going to be data collected at the network. And then even the air interface and what's exchanged between the devices and the network is also enabling more machine learning. So what's exciting about private networks is that you can have a more localized, optimized deployment. And it kind of comes down to the perspective of what are we designing for? Are we designing for reliability? Are we designing for data rate? Are we designing for latency and reliability for some types of traffic like AGV? And we're designing for high data rate transfer for virtual reality. And so the ability to kind of have different types of traffic classes, but in a private network where the deployment is, in some sense, fine-tuned towards the specific needs of that environment. I think when we look at deriving the value of these deep enterprise scenarios, then a lot of it comes down to specifics. So we can look at use cases in terms of physical scale. What's the geographic coverage we need? Are we talking square meters? Are we talking 100 square meters? Are we talking acres and acres of land? Because it's an entire innovation zone, multiple buildings, multiple types of vehicles. And so you can have this kind of physical size. And then similarly, you can look at the applications from best effort data all the way through to high-reliable controlled robot. And so I think what's interesting that within that two-dimensional view, then private networks enable the system to dynamically adjust to what's needed. And I think it's one of the reasons that we look at private networks and public networks. And then the reality of handing in and handing out and hybrid public-private networks, fully recognizing that no matter how carefully you draw your circle, there's always going to be that cut-set aspect of things going in and out of the circle. And so because of that, one of the things we've learned is that a private network, you can't really do it in a vacuum. And it's, okay, we're talking about connectivity within this physical construct only. So trying to make sure that you're designing for things coming and going. And then that's where you get the best of both worlds, where you can have that optimized private network, and at the same time allow that interaction with the public network. Thank you, John. And John, what do you think private network, maybe it can accelerate some of the architectural innovations such as open network architecture? Yeah, I think, you know, first of all, I think it's important for people to recognize that it's not a public versus private discussion at all. They're just like the IP-based world. We have public internet, so we have private IP networks in our enterprises, and they work just fine together. You know, historically, when we talked about wireless, we didn't really have that. What we had was things like Wi-Fi on one side that was largely an independent ecosystem, and then you had cellular on the other side, which was sort of a public network. And the interworking between them was clunky and ugly. And now as we move to private 5G environments that are much more consumable by enterprises, we're now seeing the technology stack actually be able to proliferate, which is the first step in kind of building end-to-end architectures. So the first thing is, you know, don't get confused that private and public 5G are independent and competing with each other. They're actually highly complementary. The reason for that is a lot of what John just described. In a private 5G environment, much like a private IP network, it is operating to serve you, a specific customer with a specific set of optimizations and requirements. And if you, it is interesting that even two factories are not equivalent. They have different data sets, different automation focuses, they build different products. And so having a network that can be adaptable and optimized for the outcome of you as an enterprise, you as a business, you as a government is incredibly important if we're talking about a constrained specialized environment. It also turns out that, you know, the 5G networks are ultimately going to be the connectivity backbone for pretty much all distributed robotics in the world. It's very unlikely that anyone doing an autonomous thing, a robot of any flavor, an autonomous vehicle, a delivery robot, a manufacturing robot won't need an advanced connectivity structure. And so if you think about where robots are showing up, they are showing up in the public environment, long haul transportation, for instance, or robotaxis, but the vast majority of them are landing inside a private enterprise. They are building out some automation within an existing business process that's quite disruptive and will come with it is an advanced connectivity environment, which is very likely going to be based on 5G or some derivative of it as we move forward. So enterprise use cases, private 5G, incredibly important, but again, very specific to your use case, your optimization, they give you the ability to deeply embed it into your automation structures into your business model. A public network can't really do that. And the reason for it is that a public network has to serve all of the enterprise. They may have a lot of features, but I don't think we want to run a public network that's got 12 million slices on it and has, you know, a million different service level agreements that doesn't work at that kind of scale. So the second half, obviously, is if we're going to have private networks, what are the public networks good for? Well, anything over distance, anything that is interworking between enterprises. So whether it be the transportation infrastructures or the projection of telemedicine, wherever a patient is, those are best served with a public network. The interesting thing is those two worlds, as I said, are not competing. And there's a third area of focus in the industry, which is, how do we make them work together? Things like neutral host gateways are great examples of saying, hey, what if you have a private network that at a control plane can work seamlessly as an extension of many public networks? And instead of Verizon or AT&T or T-Mobile having to layer a network on top of your infrastructure, the devices in it can actually interwork and share information and seamlessly move between that private environment and the public network that allows it to extend. And so I actually think this is going to be the first year where we have not just public and private, but a big focus on interworking between them, making them operate seamlessly. For us, look, you take a device, I was talking to someone who's running the logistics infrastructure, if one of the largest retailers in the world. And we were talking about robotics and the use of 5G, and they were saying, well, I'm going to use Wi-Fi initially, and then I'll move to 5G. And I'm wondering why I want 5G in my robots. And at first, they couldn't quite figure it out. And then I pointed out one simple thing, your robot might leave your factory. It might go outside. And if it does, it's going to need to roam on to a public network in a seamless way to continue to have its experience persist. But it doesn't mean you don't want a private 5G environment. You can optimize, it just means that you want to have that superset effect of being able to seamlessly interwork with the public networks also. So I think it's all of the above. It's private networks for high degrees of optimization, ownership and specialization, public networks to give you macro-level coverage and the ability to move out of your private environment and to federate with other private environments. And then a whole big focus on making that interworking as seamless as possible so it's not a radical shift as you move from public to private or from private to private as an end user. So it's going to be a pretty interesting space. But by the way, if that sounds familiar, that's exactly how the internet works. So we should all be very comfortable with that end state. Thank you, John. And by the way, after Asha and Yon, I'll give the mic to the audience. So you might want to think about the questions you would like to ask. Asha, please. Thank you. I just will add on to what John and John already said, is it's about context, right? It's not necessarily about 5G, Wi-Fi, 6G and all, but context. Just like when I am at home, I have a separate context around it and there's a lot around context, around trade secrets or competition and all in addition to things like latency and all. Today, most of the data that is generated locally is not used and not to mention the cost of taking that data all the way outside of the prem. So what I see is when I go to work, there's a context there, right? And the environment around the things that will help me with that context make a better place. So that's how I see it. And that's the main additional ones I wanted to add, right? Using the data that we generate in a way that is helpful right then and there. Thank you, Asha and Yon, please. All right. Well, John, John and Asha said many clever things. There's not so much to add, but I have one maybe, handled to this to add. I mean, of course, first of all, I do agree that it's going to be a lot of private networks to cater for the security, for the performance requirements and for the specific context of your particular business. We are selling private networks on the dozen right now, so it's big. But it's also to remember that some of these aspects, I mean, why are the public networks with this functionality gradually catch up with some of the needs, if it's like privacy or the security or when using network slices, etc. So there will be a closing of that gap to cater for that specific needs you have as an enterprise from the public networks too, that I don't say will displace private network, but it will maybe displace some of this or move some of the services that you need a private network for today to a public network who remembers the private telephone exchanges we had at enterprises, right? I mean, we thought they were there forever. But of course, after some time the public providers could or wireless providers could provide that functionality anyway, and you still have the same functionality. So with network slicing, with other aspects of reliability, etc., public networks will catch up on some of the things or cover some of the things that you need a private network for today. So I think we heard from some of the speakers earlier, right? There will be a combination here. And I think that's a good thing. Thank you all. Well, I still have one last question to ask, but first, let's see if there are questions from the audience. Yes, please. Yeah, hi, this is Frank Leonard. And I had two questions for Asha. The first is, thank you for bringing up the state of Indiana. You know, business is primarily driven locally. And I know that you guys recently announced a partnership with Indiana based scale computing. And the state of Indiana has actually got some equity in scale computing. And they're an edge computing company. And I think that the partnership is kind of geared towards AI in particular. So if you know anything about that partnership and could elaborate, it's just happened recently at a big Intel meeting, I believe. And then my second question was, could you elaborate on the chip shortage and any steps Intel is doing to help make that less painful? It seems like that's having a huge impact on the economy. And the quicker we can fix the chip shortage, I think the faster the economy can grow and avoid a recession. You did it by be muted. Yeah, I'll answer the second question. First, with the chip shortage, and essentially, you might have heard, we're building a fab in Ohio, right, and manufacturing in the West. So what more can we do than to actually increase the capacity for it, right? Our CEO recently gave a thing at Davos. And I think I'll let the details be there since this focuses more on 5G and 460. On the Indiana super computing, unfortunately, I'm not in the details or I can't comment on it. So sorry to disappoint on that. But there's a lot of work on your first question around the chip shortage that our CEO directly has very publicly communicated on various forums, including at Davos. So I'd encourage that. So thank you for the questions and the interest in. Okay, thank you for your reply. Thanks for being part of this event. Thank you. And we hope that all of the companies here on this panel will come to Indiana. And we have great talents here. And any other questions? There are two hands over here, please. Thank you. My name is Amy Glenn. I'm the Associate Director of the Office of Future Engineers here at Purdue University. I'm just curious, because I have heard that there is some concerns over health risk of the radiation from 5G. So could someone address that just to clarify? Thank you. Thanks. Anyone would like to take on that question? Radiation health concerns? I can also talk about it, but we've been around for a long time. The global health standards is set around how much the radiation warms up your body. And that's what we are measuring all the time. And the limits are set far below any health effects. And everything you have to check is the power level and the frequency use. And that's been investigated up and down for ages. There's nothing different with 5G than any other technologies. You have to know, I mean, the regulators do know which frequency bands and which power levels are sent out from. So this is unfortunately worries. And the worries we must take for real. But the physics is pretty obvious. Thank you. There's another question in the back. Hello. So my name is Abhi and I'm a graduate student here at Purdue. And I had sort of a general question about sort of how I guess 5G would be integrated into sort of physical processes. So for example, here at Purdue, there's a lot of people working on manufacturing and design. And one of the questions I have is how you could sort of integrate 5G into minimizing variability and manufacturing. Because in a lot of cases, physical material is being wasted in many of these processes. And one of the questions I have is sort of how could we sort of integrate these two spaces together? Yeah. Anyone would like to take that one? Yeah, maybe I can make a few initial comments and others should chime in as well. So I think it's a very good point to bring up variability. Because we often, it's like when you focus on latency, you also need to discuss jitter, right? And so when we look at controlling things, you know, whether you're talking as a graduate student in terms of the control theory side, or even then what is the overall operation that we're trying to control? I think what's interesting is that we look at this perspective of, okay, what is the target, you know, bound, if you will. So you can have a latency bounded control system where sometimes you're opportunistically better than that, but you're doing it to a target latency. And even this ability that as we look at quality of service, we can bring in specific thresholds that we're trying to guarantee. So similar to, you know, John from Dell's comments about, you know, quality of service and how you can look at, you know, is there a guaranteed latency or is there a whole bunch of SLA service layer agreements? One of the capabilities of the 5G system on license spectrum is that you can have targeted performance goals that you are going to meet then, based on that wireless deployment. So in terms of, you know, if you're trying to increase yield in a manufacturing sense by removing defects or even some of that jitter. So I wanted to add to that. So one of the points I was curious about is, you know, like how there's different stages of manufacturing. So for example, processes may be distributed throughout the country, right? So maybe you might be manufacturing a powder somewhere, then you might be putting together and assembling it somewhere else. So the question is how do you sort of link distributed processes that are interrelated, but each part doesn't see each other directly, I guess, in a sense. Yeah. Yeah. And I think that's a little bit how you block and tackle the problem, right? So as Meng would say, it's like a decomposition problem. So you can look at from a hierarchical framework, if you're trying to manage to some overall supply chain resiliency, then you can look at each block separately. And then you can look at these various critical paths that where you need to be, you know, jointly optimizing things. I do think it's a good question to make sure that macro perspective is incorporated, so that we're not just, you know, focusing on each problem entirely separable, but rather bringing in a kind of broader framework. And I would like to add, since we do do a lot of manufacturing at Intel, right? And factories that it's also context aware that like, you know, a lot of it is so digitally represented, like when you're talking about like the latest in process technologies, and it's very different to look at those waifers and how you model them versus the assembly lines and all. So a lot of it is very well articulated by John on the decomposition problem. But also you do need to have the models on how you actually optimize in those but integrate it together, right? Not unlike anything else we do, it's a lot of systems work. And as it gets more complex, the representation and the simulations and the emulations also become a lot more important, which is all enabled with context and lower latencies and other things. Well, I'm loving this conversation reminds me of my nerdy days when I used to be an actual engineer. But given that we're at four o'clock, I'm going to ask one last question here to this panel. And it's about the spectrum and all generations wireless communications need to use some portion of the electromagnetic spectrum. And this conversation is part of a recently started tech diplomacy conversation and spectrum harmonization utilizations important foreign policy and domestic policy topic here in Discovery Park District, we own a certain slice of so-called CVRS spectrum, just like we own some of the fibers and some of edge compute. What do you think about the dynamic sharing prospects? Is the technology going to be ready in 2030 in 6G for dynamic sharing of spectrum to be prevalent across different spectrum bands? And maybe we'll start with this time with John Rose and then we'll go to John and Asha and then to John Smith, please. Yeah, I think it's already working reasonably well. I think it must happen to John Smith's comment about the previous question. Look, if you want to optimize something, one of the dimensions that you choose from is basically dedicated spectrum versus something like CVRS or an unlicensed band. The reason for that is you want predictability at the RF layer and you want to do that through governance, not necessarily through technology. And so dynamic spectrum sharing is going to become more and more important and we're already doing it reasonably well now long term. We're going to need more of it because depending on what function you're running, it's not just choosing to do all through reliable latency or to do mass machine type communication or particular profile or slice, it's choosing which spectrum you're going to use and where it comes from and what the governance is around it. So I think it's an existential thing. We don't do it well. It won't work. I think we know that. We've already done it reasonably well in 5G and I think we'll just get better at it. We should assume that it's a big piece of how the 6G ecosystem works, that spectrum becomes a tradable, consumable, and accessible component as you build out the end end behavior. Yeah, well, I agree with John and I just add to that that the spectrum sharing is happening. It's already happening in a big thing in many ways. In order to really become efficient in getting new services out, whoever invested those services must feel that they can rely on delivering that anywhere, everywhere. And so some portion of licensed and control spectrum is needed. So my sort of my takeaway would be, or my message would be, in order to get the spectrum sharing market to take off good, we also need to make sure that whoever's spectrum has something to rely upon in the base, meaning licensed piece of some kind, because then you can spectrum share and do carrier aggregation across your licensed band and your unlicensed band or the CVRS band. So then, if we do that, then people will dare to build services for unlicensed and for shared spectrum. So that's my point. What I'd add is also in addition to a green, we also need to use spectrum better and we're doing a lot around it like MIMO or Beam Farming or other things. So we can do a lot more with the spectrum we used to have. And so all models are good, like the way I like to say is it's like land, right? It's precious. It's the essence of everything we do. So all spectrum is good. We'll use it. We'll optimize on it. And I look at it as the foundation for what it is. So and I'm very pleased to see that shared models as a way forward in addition to the whole unlicensed unlicensed because it innerweights and allows for different users, including what Mongia and the Indiana campuses enjoying already, right? Absolutely. And John, you have the last word. Yeah. And I'll just add in that what's interesting too about the sharing is bringing in the spatial dimension, the ability to share through beamforming is also very, very relevant. For example, bands like 37 gigahertz, 39 gigahertz, even those are even shared, licensed shared bands. And so the ability for spectrum sharing to open up new paradigms for how things are deployed, how multiple operators collaborate, how you can make the best use of spectrum in a spatial environment and allow reuse in another environment. I think that's really an important area of continued research and it's already happening now, you know, as we speak as part of the standards and then also more and more in these deployments. Thank you so much. I would like to virtually shake the hands of all the panelists. Thank you for your time today joining us. Big round of applause for our panelists. It has 5G evolves into 6G. We hope every step of the way the latest technology get to be deployed here in Discovery Park District commercially first. And we hope that all of your companies will consider us as a very strong partner. Well, this is also the last minute of the whole day here in DP District. And I want to thank again all the organizers. There's a lot of work and moving those autonomous combines and tractor and trailer and those zones of different sizes that can dance together. It's no small task. That's the best part perhaps of this connected community that it is a community. It takes the whole village. Thank you so much and thank you for joining us. I hope you go home. Perhaps your home is now far from here in this community thinking, knowing that here at Purdue University the cities get smarter faster in the heartland. Thank you so much everyone.