My guest today is Tyler Reid, co-founder and CTO of Xona, a company building the first commercial satellite navigation system.
We get into why Tyler and his team are moving satellites into low Earth orbit and what that unlocks. Stronger signals that can penetrate indoors, more resilient timing infrastructure, and better security against jamming and spoofing.
GPS sits 20,000 kilometres out. Xona sits at 1,100, with signals around 100 times stronger and a planned constellation of 258 satellites. Tyler came at this from the autonomous vehicle world at Ford, where the problem was simple enough: ten meters gets you to the store, but it doesn’t keep a car in its lane.
We also talk about time, and how much of the world quietly depends on GPS to keep its clocks honest. Why countries are suddenly so interested in owning their own infrastructure. And the question Xona gets asked constantly: if you’re broadcasting that close to GPS, aren’t you the jamming problem?
If you’re interested in what the future of GNSS might look like, you’re really going to enjoy this one.
More at xonaspace.com, or reach out to Tyler on LinkedIn.
In Conversation
Building the First Commercial Satellite Navigation System
Daniel: Tyler, co-founder and CTO of Xona, thanks very much for being here. Let’s dive right in. Who are you and what do you do?
Tyler: I’m Tyler Reid, co-founder and CTO of Xona. At Xona, we’re building the first commercial satellite navigation system. That means bringing new features that just aren’t available to GNSS users today. We all love GPS, this ubiquitous technology that’s made navigation accessible to billions of people. But as it gets adopted in more places, the commercial world is leading the way in wanting new features and new capability, and we’re building a system to close that gap.
Daniel: So what is the opportunity here? Where is the current GPS and GNSS system failing us, or where could it be better?
Tyler: Part of the motivation came from my time working on autonomous vehicles at Ford, on everything from highway-assist to robo-taxis. I came into that straight out of an aviation background, working on how GPS lands airplanes through the WAAS system. In aviation you need tens of meters of protection level to get an airplane on the ground safely. But cars on the road are in a much harder environment, with buildings interfering with signals, and they need centimeter-level position just to stay in their lane. These vehicles had a million dollars of sensors bolted onto a thirty-thousand-dollar car, and that just wasn’t going to scale. Lidar, radar and cameras were being pressed into localization duty alongside GNSS, and even together they weren’t quite hitting the sub-ten-centimeter, high-reliability mark these systems need. That got me thinking about investing in new infrastructure to make autonomy scalable, rather than building robo-taxis that end up costing more than an Uber.
Daniel: And space is just really cool, right?
Tyler: Space is of course cool, and part of it came from the fact that my co-founders and I met in the aerospace department at Stanford, so we’re space-minded people. But we also saw an unprecedented wave of investment in new space infrastructure right as we were starting the company. Satellite navigation had always been dominated by government, and that shift from government-dominated to commercial-dominated infrastructure hadn’t happened yet in navigation the way it had in communications or Earth observation. When I first started looking at this problem back in 2016 as a grad student, Starlink and OneWeb had just been announced but nothing had launched yet. The biggest constellations at the time were GPS and Iridium, at 30 and 60 satellites. Now Starlink alone launches 60 satellites per rocket and has 10,000 in orbit. Back in 2016 there were about 1,500 operational satellites total worldwide, a number that had grown roughly linearly since the 1950s. Now we’re past 10,000, and it’s mostly commercial, mostly in LEO. It’s a genuinely unprecedented time.
Why Low Earth Orbit Changes the Precision Game
Daniel: Your pitch, as I understand it, is more precise navigation for autonomous vehicles without needing the same pile of sensors we rely on today. Am I on the right track?
Tyler: Definitely. We categorize what these vehicles need into three things: improved precision, going from meter-level, which is fine for humans, down to centimeter-level, which is what machines need; more resilience, meaning it works in more places more of the time, like under tree canopy or in urban environments with limited satellite visibility; and security, since a spoofing vulnerability on the GPS component of a vehicle is hard for safety engineers to live with. Bringing authentication and encryption to the commercial case eases a lot of that tension. Interestingly, autonomy probably won’t be the earliest adopter of this, since autonomy itself is still finding its way into the world. What we found is a much broader set of use cases across the GNSS community looking for higher precision, resilience, or security, or some combination of the three.
Daniel: Let’s start with precision. Is this simply achieved by moving into low Earth orbit, where the current GNSS satellites sit something like 20,000 kilometers away?
Tyler: You have it exactly right. GPS today sits about 20,000 kilometers out, in medium Earth orbit, along with Galileo, BeiDou and GLONASS. We’re about 20 times closer, at around 1,100 kilometers altitude, and that changes the visibility geometry in our favor. Because we move quickly across the sky, there’s much more geometric information than you get from a GPS satellite that sits in roughly the same spot for hours. A LEO satellite scans across the whole sky in a few minutes, and that richer geometry is what you can leverage for higher precision. We’re not reinventing the wheel — we’re using the same techniques as high-precision GNSS today, things like precise point positioning and ambiguity resolution, just applying them in a new way from LEO.
258 Satellites and a Signal 100 Times Stronger
Daniel: What does that mean for constellation size? GPS runs about 24 satellites. If yours are lower and faster, how many do you need to cover everywhere, all the time?
Tyler: The ratio works out to about 10 to 1. Because the footprint of a LEO satellite is roughly 10 times smaller than a MEO satellite 20,000 kilometers away, you need roughly 10 times as many to get the same coverage. GPS’s baseline is around 24 satellites; our final constellation is about 258. That’s the trade-off of LEO — you need more satellites to do the job — but we’re ultimately aiming for similar visibility to a GPS or Galileo constellation.
Daniel: Does simply moving into low Earth orbit guarantee a stronger signal, or are you doing something else? I assume you’re not just a Wi-Fi booster for existing GNSS — this is your own signal.
Tyler: There are two factors. Moving the same transmitter from MEO into LEO gets you almost a factor of 10 increase in signal power on its own. To reach our target of 100 times stronger, we also turn up the power at the transmitter. Put together, that means a 3-kilowatt class satellite, more than 250 kilograms — not a CubeSat, but something sizable enough to put that much power on the ground for the end user.
Daniel: What about penetrating buildings, not just tree canopy?
Tyler: We operate in L-band, just like GNSS today, so we’re subject to the same physics. A rough rule of thumb is about 10 dB of signal loss per wall, so getting into a typical office building takes something like 10 to 20 dB of extra margin, which was always our target. Being in L-band also means we’re easy to adopt into existing user equipment — the same receiver hardware can often be used with a firmware or software update. We’ve seen partners take eight-year-old receivers, update the firmware, and have them using our satellite today. This isn’t theoretical — it’s happening with real products.
Daniel: So this wouldn’t be a complete solution for indoor navigation, more the overlap zone near walls?
Tyler: We think of ourselves as one more tool in the toolbox. We test indoor capability constantly with the one satellite we have in orbit today, and we’ve consistently penetrated indoors across several different office buildings, not just one. We can even calculate a position from a single satellite pass, in tens of seconds, by using the Doppler shift from its fast motion — normally you’d need four or more satellites for that. We’re still stress-testing where it does and doesn’t work. A bunker underground, probably not. A typical office building, the results so far are very positive. Once we have the full constellation rather than one satellite, the odds that at least one signal penetrates a given spot in a building only go up.
Jamming, Spoofing, and Signal Authentication
Daniel: Is the 100x signal strength also what helps with security and spoofing, because it’s harder to overpower?
Tyler: The power mainly helps counter jamming. If someone today can deny service in a small area with a handheld jammer, a much stronger signal forces them to move up to a backpack-sized jammer for the same effect, and then to a truck-sized jammer to go further. It’s always a cat-and-mouse game, but making GNSS harder to jam shifts the burden elsewhere. We’ve seen a lot of interest in this from a defense perspective — we started the company before the war in Ukraine, and the sheer power advantage brings real anti-jam capability. The security piece comes from authentication and encryption. Civilian signals have military counterparts with a security layer, things like M-code or PRS on Galileo. By incorporating something similar into our signal, we can guarantee it actually came from Xona.
Daniel: So if I grabbed an off-the-shelf GNSS receiver, I could pick up your signal, but I’d need something extra to use the authentication side of things?
Tyler: The goal is for a Xona-capable receiver to handle all of that for you automatically. About a dozen receiver companies have already implemented this, and their receivers track and acquire the satellites and do the authentication as part of the package.
Daniel: Could this reach cell phones and mobile devices?
Tyler: Absolutely — we’re targeting every GNSS user to eventually benefit from this.
Time: The Overlooked Infrastructure Dependency
Daniel: We’ve talked a lot about navigation. What about time?
Tyler: Time is likely to be one of the earliest adopters, because it has a strong need for resilience and doesn’t require the full constellation — even a handful of satellites can deliver a lot of the benefit. Take 5G microcells or cell tower infrastructure in a place like Finland, near the Russian border, which experiences GPS jamming. Those towers have holdover clocks, but still rely on GNSS to periodically reset them. A ping from a Xona satellite a few times a day can add real resilience there.
Daniel: Can you explain what a holdover clock is?
Tyler: These systems rely on knowing precisely what time it is so they can share network capacity efficiently — this person’s message uses this time slot, that person’s uses that one. They carry an onboard clock, usually a crystal oscillator, the same inexpensive kind you’d find in a phone or a wristwatch. It keeps decent time, but if you need accuracy below milliseconds or microseconds, it drifts, and you need a nanosecond-level reference to periodically reset it. Today that’s almost always GPS, because it’s available and cheap. But as jamming and spoofing threats grow, bringing resilience through new space infrastructure is more attractive than making every cell tower more expensive with higher-quality onboard clocks. It was eye-opening for us too, honestly — how much of the world’s infrastructure, financial transactions, telecom, data centers, the power grid, relies on GPS for time, and therefore on GNSS more broadly.
Sovereignty, Satellite Lifespans, and Navigating in Space
Daniel: I’ve heard people talk about growing interest from countries wanting to own their own infrastructure, as the world feels more unstable. Have you seen that interest around navigation and time?
Tyler: Fundamentally, Galileo, BeiDou and GLONASS exist because GPS exists — they’re basically copies of each other, sharing the same frequencies, signals and power. The reason they exist is politics. Historically only countries with the GDP to fund a $10 billion, ten-year build, followed by roughly a billion dollars a year to operate it, could get into that game. As space and sovereignty become bigger considerations, we’re seeing countries get creative about a sliding scale of sovereignty. We’re already working with two countries on distributing a different source of time than what’s historically come from GNSS constellations, since some countries have a realization of UTC — essentially a lab full of atomic clocks — but no GNSS constellation of their own. One project we announced recently is with the National Research Council of Canada, distributing their UTC realization to infrastructure across Canada via our satellites. There’s no reason we couldn’t do the same for other countries, and we’re working toward that with others too.
Daniel: The reason these constellations exist is that everyone’s afraid someone might turn theirs off. We have to treat this as critical infrastructure — invisible, but critical. The current GNSS satellites were designed a long time ago. Being in LEO, is a shorter satellite lifespan an advantage or disadvantage for you?
Tyler: It’s interesting to compare traditional aerospace with this newer LEO approach. Some GPS satellites are literally among the oldest operating satellites in space, older than many members of our team, over 20 years old. That’s a testament to incredible engineering — medium Earth orbit is a genuinely harsh radiation environment, and you can’t use off-the-shelf parts there. But it also means the infrastructure is a generation, maybe two, old. What we use as GPS today is pretty similar to what it was 20 or 30 years ago. That raises a real design question: do you build a satellite meant to last 10 or 20 years, or one that’s intentionally a five-year satellite so new capability comes online more often? We’re more in that second camp, the one you’d associate with Starlink.
Daniel: Could these systems eventually help objects in space navigate too?
Tyler: We think so, and people have already approached us about using our signal in space. An Earth observation satellite needs to know precisely where it is in orbit to know exactly what a picture shows. A telecom constellation needs a shared time source to synchronize satellites, the same problem we discussed with cell towers. Both rely on GNSS today because it’s far enough away, 20,000 kilometers, to cover the space segment too. But conflicts on the ground leak into space: if you fly over Ukraine today, you probably don’t have GPS, and we don’t either. We sit at a relatively high point in LEO, around 1,100 kilometers, while Earth observation and communications satellites want to be as close to Earth as possible, around 500 to 600 kilometers, for resolution and lower latency. That gap means we can bring real resilience to those operators from where we sit.
Protecting the GPS Spectrum, and Getting Hardware Into the World
Daniel: With a signal 100 times stronger sitting near GPS frequencies, isn’t there a risk you become the interference problem yourselves?
Tyler: It’s a question we’ve gotten thousands of times, and rightly so — there’s real history here, cases like Ligado and LightSquared causing interference concerns for GNSS. Coming from a GPS background ourselves, we approached this by protecting GPS and the spectrum first. When we started, the consensus was that we’d never be able to operate anywhere near those frequencies, because GPS, Galileo, BeiDou and GLONASS are already packed on top of each other, and no commercial company had ever managed it. What we did differently was avoid putting a copy of the GPS signal directly on top of existing center frequencies — we found space adjacent to them — and we used modulation techniques the original GPS designers in the 1970s didn’t have access to, drawing on decades of telecom research into fitting more signals into the same bandwidth. We ended up with a signal that’s bandwidth-efficient and familiar enough in form and function to integrate easily, but that doesn’t step on the existing systems. The hard part wasn’t just the engineering, it was the years of international coordination to prove that to everyone else.
Daniel: If I’m a hardware or software developer, how do I actually integrate this?
Tyler: The easiest path is a receiver from one of our partners. We deliberately didn’t try to build satellites and receivers ourselves — there’s already a healthy receiver industry serving GPS. As our CEO likes to say, everything from F-22s to dog collars uses GPS, and no single chip can serve every one of those markets. We’ve now integrated with more than a dozen receiver partners, spanning smartphones to civil aviation, many of them through our verification program, which gives developers tools to test their integration and confirms a piece of hardware is fit for service.
Daniel: As an end user, would I even know I’m using your signal versus GPS?
Tyler: It depends, like so many things in geospatial. Some users, think IoT devices, just want a reliable position and don’t care where it comes from. Others, like military users, care a great deal about the source and might treat our signal as a trusted backup to GPS in specific scenarios. Because that range of use cases is so broad, we lean on our partners, who understand their own customers, to calibrate that.
A Century of Navigation Precision, and Where It’s Headed
Daniel: I sketched a little diagram while you were talking, about the overlap between outdoor and indoor navigation, and eventually machine-to-human. Where does this end? Is millimeter precision the ceiling, or where is this all going?
Tyler: I’ve spent maybe too much time thinking about this. Looking at the last hundred years, navigation has improved by roughly an order of magnitude each generation, almost a straight line on a graph. In the early 1900s, mariners used sextants and marine chronometers and got kilometer-level accuracy, which felt miraculous in the middle of the ocean. Aviation came next, still using sextants for long-haul flights, but precision landing needed something better, which is where radio towers came in, delivering hundreds of meters of accuracy. A generation later, the first satellites supported things like Cold War ballistic missiles, needing tens of meters. Then GPS arrived in the 1990s, a military system at heart, delivering meter-level accuracy driven by the famous requirement to drop five bombs in the same hole. Now we’re at the cusp of the next order of magnitude: decimeter or better. Human-machine interaction, autonomous drones and vehicles, is what’s driving demand for ten-centimeter-or-better positioning, because while ten meters is plenty to help a person find a store, it’s not precise enough for a machine that needs to stop in an exact spot. Beyond that, centimeter or millimeter precision starts to matter for things like robotic surgery and augmented reality. No matter how far we get, someone’s always pushing for the next order of magnitude.
Daniel: It’s not just a robot needing to find its way to a store. That robot is probably part of a network, whether that’s a human on the other end or another robot, and they need to know where they are relative to each other too.
Tyler: Exactly. I think we’re at a transition point. GPS helped almost everyone know where they are in the world. We’re moving toward a place where almost everything knows where it is; using AirTags to find car keys is an early sign of that. That ultimately leads to a much richer, geospatial understanding of how we use the physical world: how things move through it, and why some spaces never get used at all.
Daniel: And not just where things are, but how they move from one place to another. We’re not dealing with a bunch of static objects anymore.
Tyler: Absolutely. How things move through the world is the really interesting part.
Daniel: Tyler, this is a great place to round things off. Thank you so much for your time. Before I let you go, where can people learn more about what you’re doing?
Tyler: Thanks for having me, Daniel, it’s been a pleasure and I’m a big fan of the show. If you want to learn more about Xona, check out xonaspace.com, or feel free to reach out to me on LinkedIn.
