 There is a noontime folks, Ted Rawson here, downtown Honolulu Studios, ThinkTec Hawaii with our show, Where the Drone Leads, talking about a fascinating subject often completely bypassing most people's knowledge of drones. Of course, we have a drone here on the table because we have a show where the drone leads. The guest, Dustin Tellwig, CEO of Chesapeake Technologies, Inc., out of Colorado. CTO, you were right the first time. Okay. Good memory. All right. We have a second-time flyer on this show, and once you've been here twice, it's sort of like a, you know, it's going to happen more. It works for me. Especially when you come out here all the time talking to the folks on the other side of the island here. Anyway, we're talking about spectrum, something that is really unknown to most people who deal in radios or Wi-Fi objects. The whole sinew that ties things together is spectrum, electromagnetic spectrum we should be specific on, and that's what you guys do. You guys do spectrum for a living. Absolutely. Tell us about CTI. We'll do. So Chesapeake Technology, we work in spectrum and electronic warfare and signals intelligence and those kinds of areas, primarily in the defense and intelligence arena. But you know, we all use spectrum every day. We don't see it. Every time you use your smartphone or you open the garage or, you know, in some cases change the channel on your TV, although that's infrared as well. You use spectrum, and it can be interfered with. It can cause disruption, and especially when you talk about disruption. Absolutely, if that's your intent. And you know, with things like drones, a lot of them are remote controlled through RF spectrum. And so, you know, as we use more of these in an urban environment or in any environment, they're likely to cause interference with each other and with the other things that we use that utilize spectrum. So we got to take special attention to that, figure out how to model it, how to understand it, how to forecast it, and not prepare for it. And then of course the FCC is out there managing who owns and who has access to what spectrum. And drones are going to get their own spectrum at some point in time. We don't know when. Right now they're basically running on the unlicensed open frequencies. So if we have one drone per neighborhood right now, and we've got people on their cell phones and Wi-Fi and garage doors coming up and down and this sort of thing, that's loading up that spectrum. Absolutely. If we put 500 of these in some neighborhood, or 5,000, what's going to happen? Well, they may start falling. Okay. Hopefully not. Hopefully they would just stay in place and become inoperable. But a good example, many of you may have tried to use a 2.4 gigahertz cordless phone in your house. You turn the microwave on and what happens? You hear a lot of interference and you can't talk through it. So the same thing will happen with these. There's devices we use everywhere that create noise in the environment. And as we load up, like you say, with more and more drones, they'll interfere with each other. And those normal household devices can interfere with them as well. So, you know, while we may want one of these things to deliver a package to our door from Amazon, it may become unlikely in highly dense urban environments unless we plan well. And so what's going to happen is if that density of signals is too high, the system simply can't connect the ground controller with the unit and it just won't start up, won't go anywhere. Or you'd have to put more autonomy on this. But you know, the computers that you can fit in there aren't very large. And there's a lot of things they'd have to deal with and figure out that they might not be able to. Or you'd have to put a more powerful transmitter on and try to dominate and get there before somebody else does. Which will just exacerbate the problem. Yeah. Make it work, right? Yeah. That's like saying, let's solve traffic by putting big trucks on the road. Okay, right. So we're heading for something that nobody's quite prepared to really think hard about. And because spectrum can't be seen, somebody else takes care of it. It's a very arcane subject. It disappears from our current mentality. But you've got some techniques and tools for understanding that and forecasting and predicting. We've developed a number of mathematical capabilities, algorithmic capabilities in software that can predict where that density might become problematic. It can monitor in real time if you've got sensors that you can access that show you what's going on in spectrum, what's going on in the navigation space, what's going on in the communication space. Then we can show visually where problems may occur or where they are occurring and then you can deal with them. So let's take our University of Hawaii Unmanned Air Systems test range as an example. We use our own athletic facilities. We can use our agricultural fields. We have the entire island of Lanai available to us. So as we get into serious testing in those areas and we need to understand what's out there in terms of spectrum utility and usage today, you could provide something to us that would give us that picture and make sure that what we do isn't going to follow somebody else up in the process. And what would be nice is to try that. Let's model it, see what kind of answers we come up with and determine where interference might occur and then let's go try it. Maybe not in the urban area yet, but out in the outlying areas. Let's load up the environment with lots of these things and see if it matches the prediction and then we'll find the models if we need to based on that. That's cool. I know we could actually think of what's going to happen in downtown Honolulu at some point in time when a lot of drones start showing up. They may be legal, they may be licensed and all that sort of thing but that doesn't mean that they're properly aligned with the available spectrum. Yeah and you know buildings have lots of metal in them and metal tends to interfere with spectrum as well and inhibit its propagation so we really have to have a good detailed three-dimensional and even structural model of that urban environment so we can understand how spectrum will act. Cleaner materials and structure, orientation and then that can be modeled and so the public service people who depend on radio communication for emergency responses such as they'd be very interested in making sure they have clear access to communications and not going to be manhandled by the folks out there with drones and such. Absolutely and our cell phone providers do a lot of that modeling now because it's very important to them of course to provide quality of service wherever you happen to be in the city. So you know that they've been kind of at the forefront I think of doing urban RF propagation modeling but that's one of the reasons we're interested in teaming with the University of Hawaii is because you guys have a department that focuses on some of those issues. So we could actually take that existing understanding relationship and the availability of sensors that are placed up by the cell network providers and start thinking about that as it rises up into the air and deals with drones. Absolutely. What would it take to, is that a year long process or six months? Well we have many of the tools in place to at least do the modeling and sim and some of that planning aspect of it and then you know you would want to get sensors deployed that are able to sense the spectrum that drones might use. You know a cell phone provider is going to look at 900 megahertz, 1800 megahertz the normal cell bands, drones in 5.2 gig, we might not have sensors arrayed around the city that can provide that. So we have to put sensors in place at every 1.5 mile or something like that. Wherever you know we want to model the city and we can tell you where to place the sensors. Modeling and simulation runs a world doesn't it these days? It does. You get some graphics here we can take a look at? I think so. We did a short video clip of some of our software that shows some potential interference areas of navigation systems based on some tracking data from ships and aircraft. So what you're seeing here is a multicolored heat map and the red areas kind of showing where navigation may be inhibited. So that's an area that I wouldn't want to fly a bunch of drones because they'll lose their sense of where they're at. Again this is fake data but it gives you a sense of the kind of live displays we can generate based on live sensor data and the modeling and sim that create that picture. And those red zones were the calculated overload areas where there's too much activity going on based on the simulated load of navigation devices that are attempting to communicate that you put on them. And all the way down to the modeling of the waveforms being consumed by the receiving systems in these devices. So they're experiencing lots of RF energy at their aperture and then they process that energy, try to retrieve the useful signals out of it but at some point the noise becomes so high that they can't get the useful signals anymore. Okay then at that point in time they declare I'm out of business. I give up. They either return to base or they just hover in place or hopefully not fall out of the sky. Or go to the ocean whatever it might be. You go to some kind of a default mode that is perhaps not what you wanted to run your mission. Sure. That whole issue of waveform that's something even more arcane than the pure issue of spectrum. Everything is sinusoidal these days but with software defined radios can you alter that? Absolutely. Yeah so the use of software defined radios where I can basically do arbitrary waveform generation and I can dynamically change it is powerful because that's what a lot of these devices are built on because the FPGAs on which that's done are very small but it's also a challenge for like our military brethren where a potential enemy is exploiting that commercially available capability to make very dynamic waveforms that are hard for us to predict and interact with from let's say an electronic warfare system. So you can get those software defined radios on the open market today. Absolutely. And you can prescribe your own waveform. Yep just Google Xilinx and FPGA and there's plenty of them out there for sure. And somehow you can program them to your satisfaction. Yes. If you're that type of a person you'd want to have something as far away from a sinusoidal as you can to make it less detectable. Yes there is. It looks like a very strong field. There's very complex modulation types that's a whole field in and of itself is just developing waveforms that survive well in dense environments and are unique or encoded or encrypted in ways that make it so that only your system can hear them or listen to them. Is that a possibility with again trying to jam a lot of drones into the new spectrum that is defined by RTCA whenever that occurs? Is SDR to have a lot of personalized what do they call them designer waveforms? Designer waveforms. Well in fact you know DirectTV is a good example so you know there's many many subscribers on DirectTV and there's a satellite putting a signal out but they encode their signal with unique keys that only your DirectTV receiver can decode. And so those kinds of coding techniques allow us to share spectrum in a powerful ways. So it's not always just about power level and what frequency it's in. It's how these complex waveforms mix and share. Okay and coding as well as the complexity of the waveform both can contribute to supermuxing in some way here that gets a lot of signal on the bandwidth. So we're going to see that happening here then if you're saying. It's not already. And if it's commercially available it'll be the next thing to happen here. That makes it even more difficult for the guy who's trying to detect malicious operations or someone who is like our power company out there and they are very happy with their six drones flying over the power lines but they don't want to know who that seventh one is. Yeah. What's that guy doing? Yeah. Or even a professor doing work on coastal erosion out there and someone who knows he's really good at that hey she's the second drone out there. Somebody's apparently copying my research. Yeah. Yeah or even not even put another drone up but use easily available RF emitting devices to interfere with his signal. Yeah. So the whole murky world of spectrum and waveform and the exploitation of all that is what you guys are right in the middle of. Yeah and you know again it's invisible so it's hard to attribute these things to specific individuals. When someone is interfering with your signal who's doing it well they could be a hundred miles away with a high powered interfere or they could be ten feet away with a low powered interfere. It's very hard sometimes to discern where that interference is coming from which is why we build these tools that allow us to model the environment to better understand where there's interferes. That's really great so the tools allow you to simulate and then during training you could look at actual emitters that are engaged and you could both observe operations and you could design for minimum interference and design for what you need in terms of antenna spacing power and even waveform generation in order to handle the situation that's precarious. Sure and that's really what the spectrum management community does today what we're trying to do is help them move into becoming spectrum operators so that not only are they planning well but when they're operating they're able to understand where that interference is going to occur it's unlikely that we can eliminate all of it but how do we deal with it when it does occur. That's a really interesting and complex situation we don't even have an analog in the like the road transportation system do we the roads are where they are. Well we kind of do with Google Maps right I mean the Google Maps we all trust it it gives us multiple answers on how to maneuver in that road space so we're trying to do the same kind of thing in spectrum and in electronic warfare is how do I generate autonomously solutions that give me multiple options on the battlefield. But I think you have a lot more flexibility a lot more freedom in the in the electronic world than we do on a structural mechanical world and this is again something we're going to have to educate the public a lot in the subject area as all of our systems are getting more and more dependent on wireless communication and as you start building up new buildings in the middle of Coq Aco that weren't there before then something's not going to work right you guys have a long career ahead of you here. Yeah well that's good. It's good let's talk about that after we get back from our break. We'll do. It is just afternoon folks Ted Rawlsson here downtown Honolulu Studios of ThinkTec Hawaii with our show where the drone leads Dustin Hellweigh joining us for a fascinating conversation thanks for coming back on the show Dustin and thanks Jimmy for bringing them out here. Anyway we're talking about a very hard to understand subject that we're going to all have to face our legislatures our educators our students and that is a subject of electromagnetic spectrum and wave forms that ride there in and the values and the consequences that come from perhaps somewhat of an uncontrolled approach towards usage of that air space so to speak. Yeah they do good planning today the spectrum management community does a great job but once things start happening it becomes a bit of a Wild West I think. Okay and unfortunately the Wild West also describes what's going on in the world of drones and what so we have a we have a sort of a collision sort of coming here don't we and that leads to a long career for you and your company in terms of putting up systems that model and simulate something as complex as electromagnetic radiation from radiators and what the receivers will see how the signal bounces off buildings how the signal may be absorbed by the steel in a building maybe even atmospheric effects. Absolutely. And what more do you have to add to your model as you go forward here. You can get really really really detailed and there are models that take hours if not days to run to answer one problem but what we found is you have to be able to operate and so humans need responses faster than that. So we try to create a good enough algorithms that give us a good answer that's a useful solution without having to take all that computational time in certain planning problems it might be complex enough that you have to take that time and plan it out but when you're in operations and interference is occurring you have to respond quickly. You know another thought on that I wanted to ask you drones and such are becoming smaller and smaller as power density and batteries goes up as better windings occur in the motors and such and as even radars down to you know three or four ounces these days in some case light ours down to a pound and a half as you start crunching all this stuff together we have antennas running on top of antennas. So is there within your system a way to handle that that is the scale factors on systems getting smaller? Yes. We certainly don't focus so much on what they call the close-in spectrum problem so interference on a platform itself between its multiple apertures or multiple antennas we kind of deal outside of that close-in problem space. There's a whole field called EC3 where that's what they do or I'm sorry E3 where they model that close-in interference space so that you can place the antennas and design the antennas to avoid those problems. So there are design systems and design codes and simulations that we'll deal with that the two of them are going to be needed as this administration goes over here and the civil side I mean certainly the military is very aware because they get adversely influenced by incorrect spectrum management on the civil side other than the as you mentioned the cell phone companies understand it is their bread and butter that they have to have it but the to our fire departments and police departments and emergency management people do you think there's an adequate understanding of what the potential risks might be and how those risks might be addressed out there? Hard to say probably depends on the specific department I wouldn't want to put every all of our first responders down across the country but it is a challenging field and I think some of them have probably been exposed to those problems especially if they were in the military so they understand it everyone understands when they go behind this you know certain building they lose radio contact so they what do they do they just don't go behind that building right but what if they need to go by that behind that building you know how do we how do we help them still succeed and something like a drone could help you could hover the drone at the corner of the building and it's doing a radio relay operation back into that dark space you know where they need to operate so there's lots of potential solutions there to provide dynamic coverage where it's needed. We certainly have that problem here with the high mountains the cool mountains that separate the islands in fact going up here are on the winter side of this island our television feed came from the island of Maui and I think our first responder community has a lot of that going on with a lot of a lot of signal propagation issues that the mountains bring up. I'm sure there's a mountain peak somewhere around here that has lots of antennas that are just radio repeaters between the sides. You're suggesting a radio repeater could be a small one could be whether at low frequent low power low range but that's good enough for an urban environment or even a search and rescue environment up in the complex mountains. There's one thing we can't solve it's the physics problem of wavelength which is you know a descriptor of RF energy and wavelength is keyed to the size of the antenna and I think I pointed one out here earlier. That's a dipole antenna for 2.4 megahertz. A small antenna is for higher frequencies so the higher the frequency the shorter the wavelength and therefore a 2.4 gig antenna can be very small. If you get down into 300 megahertz that's where you start seeing longer whip antennas you know and into HF frequencies which propagate very far but you need a 50-foot antenna to be able to receive them. So there's a limitation to the miniaturization it's going to have to stay in the in the ultra-high frequencies in order to keep the physical dimensions from becoming a problem. But on the negative side of that those frequencies don't propagate as well so they're easier to absorb and interfere with and they don't go as far without more power. So therefore you have a better operating environment in that regard. Yeah and you'll see the military has picked certain frequency ranges to best enable them to communicate on the battlefield so they need you know good propagation distance without too much power and so as we lose these frequencies as they get sold off by the government to commercial industry which is good it's a revenue generator the military is having to understand how to deal with that. And there's a thing called the National Spectrum Consortium that's actually been stood up to help develop technologies to solve those problems. Is that part of the Department of Commerce right alongside RTCA? I think NTIA is part of the Department of Commerce. The NSC is a it's called an other transaction agreement consortium so it's kind of like if you're familiar with DIUX or softworks these rapid acquisition conduits the NSC is something similar to that to get capability develop fast. You know the background and experience you guys have working your company has working in these complex battlefield environments with the military is really useful as the drone community starts to generate its pulling together its standards. The drone community these systems are going to have to be self certified. The FAA is not going to generate regulations they're saying you guys go sort out yourself best practices inherit them from wherever you can make them correct and then adhere to them. And if you stay within the bounds color within the lines and don't cause problems what you go that way. If you don't do that we'll come in and be glad to manage you. We'll have all the regulations we're familiar with on airplanes. So right now the industry is attempting to gather itself to get at the search and rescue people the post disaster of building inspection people the insurance you know fall away from the user all the way to the trusted chip designer on it in a trusted software system like folks dealing with multiple security and such so the whole gamut from calculation to physics is all up for having standards associated with it being put together and your and your knowledge ought to be in the middle of that. So there's a I'm actually a member of a subcommittee I probably don't attend the meetings as much as I should so I apologize to the IEEE rather out there but IEEE 1900.5.2 has developed a model called the spectrum consumption model and they're pushing that forward as a standard descriptor of what they call spectrum consumption which is transmitting or receiving actually and so we're working some programs with organizations like DARPA the Defense Advanced Research Projects Agency to leverage those standards to describe this complex modeling and highly scaled up dense environments and how do we manage that and how do we command and control in that kind of environment. So that that basis that you got you could offer it to the to the growing organizations that are coming together here. And like I said I don't need to it's already an IEEE project so that they certainly have it out there and you know IEEE is a great organization for pushing standards forward. That's great so and that really also means that basically all the things we have today that are drone are going to be cycling through some major conversion here and so what we have is going to kind of disappear and something to get replaced by a whole new family. It probably has in it different performance standards certainly has in a different fire safety and structural aspects and operating and system reliability and spectrum management. Yes absolutely you know today like you said earlier most of them leverage unlicensed frequency bands which is you know definitely Wild West you know the government basically says in this frequency range do whatever you want as long as you're below this certain power level. As we aggregate the drones even though each individual one may be you know below that unlicensed power level as we aggregate them they'll start interfering more and we got to figure out what to do with that. Okay and we would like to increase our relationship with you and at University of Hawaii in terms of our test range and the aspects that stream off from that including reaching down to the to the STEM programs in the high schools and such and we got the PCAT conference going on this week I'll be reaching there tomorrow and I'm absolutely going to bring this story to that to that table. Great and we funded a STEM scholarship with University of Hawaii through the Association of Old Crows this past year and so we certainly are supportive of STEM. If we can't get the students excited in our domain then you know we'll have a brain drain and oh guys like me will disappear and we won't have anybody take our place. I'm older and I'm disappearing also but actually we got to get them by the time they're at seventh grade right? Absolutely. If you don't get them by the time they're seventh grade they're going to disappear off. In fact part of my talk tomorrow is going to be interesting I'm going to thank the teachers I had in seventh grade for what they did for me and a lot of them didn't do it for me but we survived each other I guess but there were a few that were really significant in helping understand that but you have really certainly helped me a lot and hope our audience in terms of understanding what this totally arcane subject called spectrum is all about the thing you can't see but it's the thing that most keeps everything working together these days. Absolutely. So Dustin thanks for coming on a show next time you're out here we'll get you out again. Absolutely. Okay we'll see you next Thursday folks.