The Role of AI in 5G Technology with Dr. Tom Rondeau

[00:00:00] Tom Rondeau: The real challenge that I seem to face is everybody understands the value of what we're doing. They all know that communications and networking are key to it. Not everybody understands the difficulties of actually pulling it off.

It's wireless, uh, you don't see it, which many people equate to magic. I've often joked that to most people, I'm a wizard, the way that I can actually manipulate this magic force. Um, so it becomes out of sight, out of mind. And so you forget how critical it is to the operations of, all of your daily lives.

Until it's not there. 

[00:00:52] Ryan Connell: Hello, this is Ryan Connell with the Chief Digital and Artificial Intelligence Office, joined here today with Dr. Tom Rondeau. Tom, how you doing?

[00:01:00] Tom Rondeau: Hey, good. Thanks. Good to be here. 

[00:01:01] Ryan Connell: Yeah, hey, just, I'll toss it over to you for a quick, uh, self introduction.

[00:01:06] Tom Rondeau: Yeah. So thanks. Thanks again for allowing me to do this. This is really awesome for us to have this opportunity. So, yeah, Tom Rondo. I am the principal director for future generation wireless technologies or future G. we sit under the office of the undersecretary of defense for research and engineering.

and so my remit is all things, commercial cellular, wireless, you know, 5g research, advanced 5g, 6g, anything that's coming next. So that we as the department can make sure we're tracking and get our arms around it.

[00:01:35] Ryan Connell: Awesome, no, I appreciate that. Alright, so first of all, I didn't know 6G was a thing, so, help me understand what that is.

[00:01:40] Tom Rondeau: 6g is becoming a thing, uh, rather rapidly. it's a very interesting time for us. About every 10 years, you see another generation of cellular technologies come about, which means that as they're being deployed, we're already thinking of the next generation technology. Uh, and so we've been talking about 6G for a few years now.

I think we're actually in a critical moment of defining really what is 6G. What are we really going to push for in 6G? And we've had international conferences and communications all across the world about nailing that down. Uh, so it won't be a thing for another few years. We're still working through the standards and commercially it won't be deployed probably until, you know, 2030 ish.

but yeah, but we got to think about it early if we're going to be able to, take any leadership position in it.

[00:02:23] Ryan Connell: Wow, okay. So, uh, you, you said something before we jumped on here about, uh, hey, it's, it's not 5G like that's on your cell phone, because that's the only 5G that I know of. you want to get into that and like, what is 5G?

[00:02:34] Tom Rondeau: Yeah. Yeah. So, there's a whole lot of directions we can take with how to describe this, but let me just go direct to the question here. Right? So 5G, as it was defined, was a revolutionary idea that we could use software defined networking to manipulate all of the cellular resources. So one of the things that we use cellular services for is downloading data to our phones, right?

4G gave us an incredible data network. And so we just got addicted to the fact that anywhere we go. We've got a huge access to a huge amount of data, through our phones. 5G wanted to advance that, but it also wanted to branch out and do more things than just more, uh, what they call enhanced mobile broadband.

that's what you see on your phone today is enhanced mobile broadband, just 4G on steroids, right? Bigger, bigger pipes trying to get more data down there. but there were these other aspects to, to the network that 5G started to explore. Because instead of just pushing all of the resources to get you more bits per second, we could also push those resources to give you more guaranteed latencies on your connections.

So when you're talking about things like, telesurgery, right? Manipulating robotic arms across the world, to do surgery with somebody sitting, you know, in their office in, in, you know, say New York or DC, but manipulating this robot around the world, you can't slip, right? you can't have errors in that network. 

Using those resources to say, I want reliable and guaranteed latencies on all of my connectivity so that when I move in point A, the device moves exactly the way I want it to in point B. So that zone aspect of where 5G, the novelty of 5G comes in is to manipulate the resources to that.

another way is what we call massive machine type communications. So machine type communications is the internet of things. so all the devices that want to talk to each other, talk to the network, you know, sense information around you, provide some edge processing and give you information about the, the environments back to you.

the vision there is that you're going to have Thousands, tens of thousands, maybe even millions of connections in a very dense area, on these networks, they don't necessarily take a lot of data, right? You're not talking about high def or ultra high def movies to your phone. what you're talking about is a little bit of information being passed by a lot of sensors or Drones, if we have swarms of drones, each one of them needs a, a robust connection, but each connection is, you know, kilobits per second, maybe a little megabits per second.

so now we want to divide our resources to communicate to all of these things simultaneously. so those are different ways of being able to partition up your network. And that's what 5G was, was designed to offer us. the challenge with it is the telcos, uh, your Verizon's, AT& T's, T Mobile's.

great American companies, they know how to, uh, to sell you as a user or contract to give for a monthly service to access through your phone or your computer or your tablet or whatever. we don't really know where the markets are coming from, or they, We think we know where they're coming from.

We haven't seen them develop as efficiently or as fast as we wanted them to. So a lot of those novelty aspects of it, a lot of the cool things that I just talked about, those are still in development and we're still seeing those markets and technology areas grow to meet what the network can provide for them.

So, yeah, most people just see 5G on their phone and they say, well, is this really worth it? It's going to be worth it. Once we see all these other things come to fruition.

[00:05:51] Ryan Connell: Yeah. So, so help me understand. I mean, I think the drone example was a start here where I'm kind of following up on, but like, what kind of impact do you see this having, maybe within DOD or within national security? 

[00:06:04] Tom Rondeau: Yeah. I mean, it's, massive. Uh, we're already seeing drones in the field right now being controlled by 5g. they're not quite, you know, 5g enabled in, conflict right now, but I think it's only a matter of time. we have been working with our, partners, uh, allies over in Latvia, uh, in Latvia Mobile Telecommunications, uh, LMT.

they've actually been building drones that have 5G chips built into them and have been, exploring how to use those for military purposes. over in Taiwan, some of the companies over there have been, developing, again, 5G enabled drones. this is actually a really cool one. We were over there a few months ago, and they were showing us, they took racing drones.

So, those really small, fast, first person view drones, that usually are raced by a direct connection to the drone. But what this, these guys did in Taiwan is they actually connected that through a 5G network, but a straight 5G kind of off the shelf 5G network wasn't going to hit the latency requirements that the, pilots required.

So working within the standards, so it's still standards compliant, just a different configuration than what would be most deployed for this enhanced mobile broadband, again, back to that software defined aspect of it. They were able to manipulate the signaling quality of this so that the packets were smaller and they had less overhead so that the latency was reduced enough that the pilots could fly them at the speed that they're used to.

Meanwhile, they use the same 5G network with a different, part of the network sliced off to, manage for, uh, high definition video. And so they were broadcasting this event on ultra high def. televisions around the arena while they were projecting holograms that were also produced off the same 5G network.

So, really taking advantage of all these, capabilities that 5G brings to us so that we, so that you can now fly these drones at speed. So it's that integration of 5G capabilities into these advanced robotic systems and the ability to manipulate them. That's becoming really exciting for us to see.

And you can extend from there with what you think we think you could do with, with military capabilities, just for, from the autonomous vehicle perspective.

[00:08:04] Ryan Connell: Awesome. No, I appreciate that. Um, you'd mentioned something, uh, earlier about, uh, support to NATO. Is that something that you're working on? currently?

[00:08:12] Tom Rondeau: Yeah, absolutely. In fact, we're about to head out there,back to Europe. In fact, it's in Latvia again, working with our allies there. Uh, NATO and the Latvians have been really pushing forward on how to use 5g for interoperability and enhanced security for, for partner force operations. So.

You get together with NATO, all 32 countries having to agree on something. Uh, one of the key challenges there is an interoperability. So, everybody has their own, specific radio that they want to bring to the field. Everybody builds their tactical radios against certain, uh, security risks that they assess.

and so we have this proliferation of, non interoperable, different types of, uh, devices out there. Well, one of the things we've done with NATO is shown that using an internetworking function that commercial industry designed to interoperate with wifi. So wifi device could become part of your native 5g network.

We use that to talk to tactical radios. And so now you can actually communicate between multiple different forces, tactical radios. They're all actually communicating, the physical layers communicating to You know, within their own technical radio format, but the networking portion of it is controlled by 5g and then 5g becomes your interoperable core that everybody communicates to and communicates through.

So all the data coming from US forces operating a PRC 163 over to, you know, a Dutch or a Estonian or Latvian. or Norwegian tactical radio are now actually interoperable because the core technology allows us to merge these systems together natively within the standard. Uh, we're also working with them on authentication and security features.

So you come to the field and even though the Latvians are running the 5G network, You have a core back in your home. So say Estonia, we have a core set up in Estonia. They can actually roam back to Estonia and authenticate through their process and set up their security posture, uh, for,access.

And so now everybody kind of can come to the field, bring their technology. And we're helping, coordinate them while allowing them to bring, you know, to have their own autonomy and their own independence on what, technologies they want to use. And now we all just coordinate through this international standard that we've all agreed on.

[00:10:27] Ryan Connell: that's crazy. So, I'm just thinking even outside of, of NATO, there's A million us only, examples of things that don't talk to each other. Is that in the same realm of possible?

[00:10:39] Tom Rondeau: Yeah. So we actually really started that interoperability problem with us only radios. so taking a lot of our tactical radios, that, that exists in the field right now, we showed how they communicate to each other, and We actually patched one signal through an HF radio. So we took like a standard UHF radio and moved it through, and broadcast over an HF radio, uh, so some pretty cool stuff there just with us interoperability.

and then it was quickly identified that, Hey, this is a big problem in NATO, our partnership with them and kind of driving needs, especially in Europe right now with Russia's war in Ukraine, yeah, this, became a really exciting possibility for us to expand what we were doing with

[00:11:15] Ryan Connell: Got it. And then, um, because I'm not a technologist, I want to make sure I kind of understand,the example that you threw out, but you talked about leveraging the network. and what ran through my mind is like, and please feel free to say, no, Ryan, that's wrong. but what ran through my mind is.

Like the way that,is it Apple tag? Uh, is that what, uh, thing that you can 

put on a device? AirTag. Thank you. Yeah. And so like, I just realized how these work. I had no idea, but effectively, as long as there's any Apple device nearby or any Bluetooth device, it'll ping. It's not, it doesn't have to be your Bluetooth device.

Am I close or am I way off?

[00:11:48] Tom Rondeau: Uh, yeah, I wouldn't, I would not as well. I guess it's a little in the sense that there's heterogeneous access to the RF spectrum, so Bluetooth is one physical layer model. For moving bits across the air between, between two antennas, the five G what they call the new radio and our way for is another physical layer model for moving information across between two antennas, Wi Fi, four G.

All these are different ways of doing this. so every tactical radio that we have. Has a different way form designed by the company. And usually in the military sense, these are proprietary way forms, often. and so, from company A to company B.

Uh, even if they were on the same channel, they wouldn't necessarily be able to communicate with each other because they're just different waveforms, different ways of talking, simplify it. One's talking Bluetooth, the other one's talking Wi Fi. both operate at 2. 4 gigahertz, can't talk to each other.

What we're doing is we're creating Company A talked to company A's radio, at the base station, but at the base station, it's plugged into our core network. So the physical layer is just moving information between the antenna to antenna. the network is how you encode the data for communicating between two points.

And this is the same if it's wired or wireless. And so we're looking at the network layer as that interoperability layer. So we're translating between. These different physical layers and coordinating through the 5g core to, to manage it at that, the network interoperability layer. 

Hopefully that made sense. 

[00:13:20] Ryan Connell: no, that definitely helps. I appreciate it. And, and, you know, just like any other technology advancement, um, you know, you have, uh, police radar, then you have radar detectors, then you have wider than you have wider detectors. Uh, what's that state look like in terms of the cyber security risk with the increased, uh, interoperability?

[00:13:40] Tom Rondeau: Yeah. This actually goes back to my, earlier work when I was, running a thing called the GNU radio project. So GNU radio, GNU, it's one of the GNU free software foundation projects. and this was a free open source software platform For software to find radio. So we could download this software package connected up to some, relatively inexpensive hardware that, that friends of mine were making, and now you have very easy access to electromagnetic spectrum, both transmit and receive.

And so I always joke that I used to spend a lot of time apologizing to the FCC,specifically the old director of the office of engineering technology who loved what we were doing because we were innovating and we weren't really causing any, any havoc. But it did break open this idea that you can now manipulate the electromagnetic spectrum and you can see it in ways that we hadn't been able to observe before.

So a lot of security in cell phone or in wireless in general had been Just based on the fact that it was so hard physically and costly to get access to the spectrum that you just had that natural built in security Well, we've kind of broken that down software defined radio has really broken that down And now it's fairly trivial to get access to spectrum people will transmit and create, you know, attacks against systems or jam or cause interference of systems, as well as observe the system.

So, you know, this is that dual use or double edged sword kind of concept where you can use that for good and you can use it for bad. for good, you know, we need to have more eyes on what's happening in the spectrum. I always think of the electromagnetic spectrum as a maneuver space. it has many dimensions.

You know, we have three dimensional physical space and we move in time, between all those dimensions. There's many dimensions in the RF spectrum, and we want to be able to move, in time, frequency, space, uh, in order to take it, best advantage of that resource. and so the, the software defined radio allows us to do that by waveforms or receive them.

But you can't maneuver if you're blind. So, you know, so we need, to be aware of our surroundings and we need those sensors to help us with that. So both as a weapon to be able to attack these systems, but also as the ability to protect our systems by knowing what's happening around us, informing the systems of what's going on in the spectrum and making smart decisions are all part of the.

The future of our wireless communications.

[00:15:56] Ryan Connell: Got it. And so you're talking about some autonomous type decision making. So maybe that's a good segue to, we're talking with CDAO. So, uh, AI, there, there must be some, AI, there's an AI play in almost everything these days. So, so help me understand the 5g future G AI, uh, relationship.

[00:16:17] Tom Rondeau: Yeah, they're so tightly coupled together. and we're looking at it in two directions. There's an internal AI play and an external AI play. Internally, it's these making these smart decisions. So, as we are increasing our use of the electromagnetic spectrum, and the waveforms themselves become much more dynamic and much more flexible, so that they can create the quality of service that you require as a user, they can create the quality of service that you require as a user.

and they have to avoid things like interference or jamming, or other, forms of exploitation and attack. You need the system to be smart enough to manipulate that because what you don't want, what I don't want is my warfighter sitting down there on the front lines, having to worry about what their radio is doing at any given time.

The network should know that and be able to make smart decisions. So that's a, that's me worried about it from that, the warfare concept from a commercial concept. To what we're worried about is just. Use of the spectrum and it's a resource. It's a very expensive resource in the United States. and our ability to use it as consumers is dependent on the network's control of that asset.

And so being smart about how you manage the presence and the, and all those resources to respond to changes in the environment or changes in the operational or users needs. that's where AI is going to come into play inside the guts of the network. So making those decisions for the edge endpoints is going to be driven by a smart AI enabled, brain basically of the within there.

and there's a huge growth path for what we can do, in that scenario. Uh, real quick tangent, one of the biggest cost drivers of, of telecommunications networks is power. so anything that helps them with energy efficiency and by making smart decisions about which radios to use and which ones to turn off.

Where to do the computing or where to offload, some of the resources for the computing based on, you know, things like energy prices or, sustainable energy, uh, availability. Those are all going to be really important to the networks to decide how they improve their overall power efficiency and lower their costs.

good for us, good for people, good for the environment, but also driven by, by business interests. So that's, that's on the internal of the networks. External, what we're looking at is,is really, I'm going to actually put it into two different categories here too. The first one is, contextual learning, right?

So the next generation of, AI and how it helps us better understand our environments, our surroundings, and making smart decisions for us is going to be. generated by knowledge and sensors of what's around us. So the tie between your network and your AI system has always been coupled. First generation of AI expert systems was when you finally had a workstation that was powerful enough to compute this expert system locally for you.

Then we had the explosion of fiber and high speed access to what then became the cloud. And the cloud became where you store all your data and all the compute cycles so that the current generation of machine learning, statistical learning was made possible as you push forward into the future as edge computing and contextual information is going to be critical.

Having this kind of 5G, future G access to the networks to pull in that data and manage all of that in an efficient manner. That's all going to be part of the future of how we make sense of the world around us. But finally, on this topic, we're looking at a thing called integrated sensing and communications.

So Isaac, as we call it, is how do we use the signals around us? So all of our devices are now broadcasting RF, RF's bouncing around every, everywhere. And we have all these towers and all these receivers that are listening to this RF, uh, information. What can we learn about the world around us by looking at these signals as though they were, you know, the physical properties of them, not just the information that data contained in the waveform shape itself.

That's a really exciting possibility for detecting how many people are somewhere, you know, the activities of those people, cars, Back to the drone swarms, if we're gonna have like legit drone swarms for commercial, you know, commercial, capabilities, we're gonna need traffic laws and we're gonna need to be able to manage and uphold those traffic laws.

Well, how do you do that if you're not sensing constantly where the, what's happening with those drones, the networks that we're building could be an inherent part of that, of that model for observation and knowledge. That's really cool. And so integrated sense of communications has a lot of potential for data, for analysis and understanding of our world, which translates into some pretty cool stuff that we can do in the military since intelligence, reconnaissance and surveillance is one of our big mission sets, but it also should worry us if this is commercialized because now anybody can use this and can they use it against us?

And so we always look at how can we use and defend against,this growing concept of ISAC. Uh, and to round it out there for your question, if you're talking about all of the signals, all of this bouncing, all of the analysis of this information that's coming into these antennas, you're not going to be able to separate that or split that out without some advanced AI as part of it.

So these are all coupled together in the future of, of wireless communications.

[00:21:24] Ryan Connell: That makes a lot of sense. And you're, I think maybe part two of part four of that kind of hit on something. I was going to follow up with a question on, you know, I started thinking about as you were talking, I started thinking about, take a generic large language model that has just gone out and read, read it and, you know, Facebook or Twitter or whatever.

Right. Um, It's effectively like free data besides the cost of ingestion and compute. and you have what sounds like a, Ryan's opinion. Correct me. Like, it sounds like there's a challenge in like, you don't have that free data. You're effectively having to, Start if you haven't collecting the data with all of the transmissions between different radios in that exampleand how it made the decisions to bounce from one to the other.

And if that was the right decisions, because effectively you need data to train the AI model.

[00:22:11] Tom Rondeau: Definitely need a lot of data. And because of the, so it is a lot of data. these are massive data sets that you're going to collect them because the amount of bandwidth you're taking is proportional to how big that data set is. And this will, it will quickly swamp how much data is on Reddit by just doing by just sampling the wireless communication systems around us.

So it's a lot of data and it's a lot of processing to handle that data. So where do you store all that data? How do you get access to it? and also because that data. It has a lot of information about us as people, right? U. S. Person's information is effectively embedded in this data set. there's a lot of laws and a lot of regulations that have to be carefully considered and, respected, so that we don't impact the privacy of our people.

well, we're also be able to use it for really useful, interesting, novel applications. So it's a huge problem. One of the areas of growth right now in the wireless communications world is, digital twins. So can we build more effective simulations of wireless environments? And again, because of the amount of data and the amount of compute that's required to do this has been really difficult to do well.

but we're making, we're seeing some pretty good strides in industry right now, being able to really effectively do,do digital twinning and therefore simulate a lot of environments and collect that experience before taking it to the field.

[00:23:29] Ryan Connell: Awesome. Yeah, it sounds like a smart way to tackle it. How much, and potentially kind of segueing to a different topic, but like, how much of all of this, I'll use the word code, but like, are you, compiling and it's, I'll say proprietary or how much is it open source? Like, what does that look like? Uh, you'd mentioned earlier open source was important.

Uh, so let you, uh, touch to that. 

[00:23:52] Tom Rondeau: there's kind of a, two trends happening. the near term trend is what's called open RAN. So, uh, radio access network. This is effectively the base stations that you talk to. Your phone talks to every day is called the RAN. the RAN is about 80%. Of the cost of a network, both in the CapEx and the OpEx of running your network.

A lot of the just the equipment costs and the energy and the running of these costs. So this dominates the network performance and the network costs. And classically, the RANs have been, you know, well, in the past, uh, 15 ish years, they've really been consolidated into the HENs, the Huawei's, Ericsson's, Uh, Nokia, Samsung, and then ZTE is quite far down, on the list right now.

So you've got those like five companies that are dominating the RAN, uh, market, and three of them are really good friends of ours, Ericsson, Nokia, and Samsung, we have a great relationship with all three of them. They do amazing work. What's challenging though, is breaking those systems open because the proprietary nature of those that run our networks is becoming challenging, not only for us to run.

But for our ability to integrate with the network, for the DOD's ability to innovate on top of the network, adding our own security features, all those are kind of are restricted to us. we also think that industry is going to, the classic RAN industry will benefit from breaking these systems open because they're going to then, uh, have an explosion of talents that understands how to manage these systems.

Uh, new vendors are going to be able to come about and help them with, new capabilities and innovative solutions for certain problems. So breaking open the RAN into an open RAN, where you have these defined interfaces by an international standards body called the O RAN Alliance is currently the.

Kind of leader in this area, that's going to, that's creating a whole new marketplace,and, enabling us to operate in these networks in a different manner than before, we've been heavily invested in open RAN in my office, uh, for a number of years now, we've run open challenges, the 5g challenge and the Rick forum, all of it's based on our promotion of open RAN and trying to incentivize the market.

To create a more sustainable and more secure market of trusted vendors that we can work with. So that's where we are going right now. The leapfrog, the next, I think the revolutionary idea that we're pushing right now is an open source software model for 6G. So, what Linux did for the internet and what kubernetes did for the cloud, we want a similar open source platform for industry to be able to leverage for growth within 6G.

And if 5G advance and 6G are synonymous with advanced artificial intelligence, uh, and advanced robotics and, and automation, the ability for us to explore that and to work within, you know, at the boundaries of, the cyber and physical worlds, we need that, open platform and a shared burden of that platform, to take us forward.

So this has been coordinated across U. S. government. We've been working with many agencies and many other departments in the, in the U. S. government. We've talked to a lot of people in industry and,they're either excited about this or,or willing to jump in with us early,to help us build something like this up.

but this is going to be the game changer for our. Work on the speed of innovation, our security analysis of the code base and of our rollouts, but allow more people to be better educated and informed and work within this, construct.

[00:27:10] Ryan Connell: Yeah. So I'm trying to visualize, uh, are we talking effectively? Like. Having some version of a shared infrastructure and data layer and then letting,the platform owners and apps kind of build within that construct.

[00:27:22] Tom Rondeau: That's precisely it. Yeah. So this is not a way to,to break business. This is a way to incentivize new business and to create new opportunities for business differentiation. And instead of having to create your entire ran. Uh, yourself, if you want to innovate in one piece of it, now you can build your own ran based on this, construct and then innovate within there.

And so we look at things like applications or some of the controls and like the AI brains that are going to be, you know, I talked about earlier, manipulating these networks. those are going to be where a lot of differentiation is going to come in and a lot of, you know, ways that people are going to be able to build products and sell innovation.

Or, because a lot of the capabilities within a 5G, future G networks are incredibly computationally hard. There's been specialized chips designed to solve some of those problems before. More and more, it's going to commodity hardware, but there's still opportunity for specialized hardware to tackle very, very hard computational problems.

that's another area of competition that we would love to incentivize. But if you just have an idea for how to build the next great. Encoder decoder chip, you don't want to have to build the entire ran to do that. You want to, we can now integrate that and have the interfaces to communicate with this platform, but offload what would be a software approach now to a hardware approach, get that energy efficiency, maybe get the costs down.

And now you've again, differentiated yourself in the market.

[00:28:55] Ryan Connell: Got it. Um, it kind of a follow up, but, thinking about, you know, me and my role here and listening to you and thinking, uh, how to overcome some of the challenges and making sure seeing if there's, a relationship, but like, is there the same type of like, authority to operate process within the, I'll say like telecommunications 5g space as there is within the software space.

Or is that something that you deal with on a regular?

[00:29:18] Tom Rondeau: It is, yeah, for, for interoperating with any DoD networks, we still have to get an authority to operate. And it is, challenging when you are a software defined network. So, uh, and upgrades that software and how to make sure that you're consistent with the ATOs, uh, but it is absolutely a thing we have to deal with.

We've been researching a lot of 5g installations. We've been deploying a number of them around the country, some around the world, within us, uh, installations. and they've all been isolated networks up to now. So we've been working all those wickets, uh, for getting an ATO for us to move to that next level of true integration with the rest of the DODs enterprise, absolutely a challenge for us that, yeah.

[00:29:54] Ryan Connell: Got it. Hey, uh, kind of getting towards the end here, uh, wanted to turn it over to you and see if you had any kind of last words of wisdom for the listeners.

[00:30:02] Tom Rondeau: yeah. Uh, I wanted to just point out again, how critical I feel. I feel and pushing this type of technology, the real challenge that I seem to face is everybody understands the value of what we're doing. They all know that communications and networking are key to it. Not everybody understands the difficulties of actually pulling it off.

It's wireless, uh, you don't see it, which many people equate to magic. I've often joked that to most people, I'm a wizard, the way that I can actually manipulate this magic force. Um, so it becomes out of sight, out of mind. And so you forget how critical it is to the operations of, all of your daily lives.

Until it's not there. No one remembers the end, the network engineering until the network fails. Uh, so these are things that we have to continue to push on, continue to evolve. And understand that it's not, even though the end application of how we use the information or the, uh, You have the application that we're enabling at the end.

That's what we're here for. We're infrastructure to support that, but we're really important infrastructure. And to get there requires a lot of resources, a lot of talent, a lot of research and development left to do for us to be able to do this, uh, securely and efficiently for the rest of the department.

Um, so it's a really exciting thing. It's really exciting time. we're here to, change, uh, the Our ability to live, operate and work within this world of, wireless communications and how industry has been, running wild with it, uh, allowing us to get to where we are today, but we're nowhere near done.

[00:31:28] Ryan Connell: Awesome. Yeah. And I, I can imagine that people will realize you talked about not having it until it's gone or you miss it. but, um, when you're going into surgery and there's a robot doing your surgery. I think people will, realize the importance of 

existence. Yeah.

[00:31:41] Tom Rondeau: you got to trust and rely on that network. Yeah, 

[00:31:43] Ryan Connell: Yeah.

Awesome. Hey, Tom, appreciate you being on today.

This was a, this is a real fun experience. thanks for being here.

[00:31:50] Tom Rondeau: Good. Really appreciate the, uh, the opportunity.

[00:31:52] Ryan Connell: Awesome. Talk soon.