AIN Roundtable—Readying for Potential Navigation Spoofing, Jamming

GPS spoofing and jamming—and more broadly, interference with global navigation satellite systems (GNSS)—remain a security and safety concern for operators worldwide. While technology companies are looking at a range of solutions in the flight deck, regulators and industry experts are looking at operational guidance to arm pilots for these encounters. AIN convened a panel of experts to discuss the problems that continue to plague operators and fixes on the horizon. Honeywell Aerospace sponsored this roundtable.

The Participants

Ben Mohr—Offering Director for Alternative Navigation Products for Honeywell Aerospace

Mohr’s experience with Honeywell Aerospace spans 22 years, holding different roles from research scientist to manager in the navigation field. In his current capacity, he is overseeing product lines to work in GPS-denied or degraded environments for the aerospace giant that recently spun off into its own company. Honeywell Aerospace produces a range of navigation and sensor systems for the business aviation, commercial, defense, and space sectors.

Jiri Marousek—President and CEO of AirTera

Marousek brings more than two decades of experience building, acquiring, integrating, and scaling technology across aviation, defense, compliance, biometrics, and enterprise SaaS. Formerly NATA Compliance Services, AirTera has brought together technology companies in the safety and security arena to offer services from safety management systems, data feeds for flight risk assessment tools (FRATs), identity monitoring, and other programs.

Serge Christiaans—Aviation Cyber-hybrid Defense Specialist for Dyami Security

Christiaans, an Airbus A320 captain and military veteran, also provides cyber-hybrid defense expertise for Dyami Security, a global security firm based in the Netherlands. A former chief information security officer for the Asia-Pacific region who continues to fly Airbus jets for various European airlines, he founded the Aviation Cyber Academy in Singapore two years ago to train commercial pilots to recognize and manage cyber and hybrid threats in flight. A regular speaker at global forums, he further provides consulting services on cybersecurity and hybrid security issues.

Richard Boll—Chair of the NBAA Domestic Operations Committee’s Airspace, Air Traffic, and Flight Technologies Subcommittee and Retired Corporate Pilot

Richard Boll recently retired as a corporate pilot who flew Bombardier Learjet 75s and Challenger 300s and 350s for more than 40 years. Working with NBAA for the past 20 years, he remains active with the association, chairing the airspace, air traffic, and flight technologies subcommittee of the Domestic Operations Committee. He also represents NBAA on various FAA and industry forums, including the Performance Based Operations Aviation Rulemaking Committee (PARC), the Navigation Working Group, and the GPS Disruption Action Team (GDAT).

The Discussion

On Explaining GNSS Jamming and Spoofing

Ben Mohr: We talk mostly about GPS, but these threats certainly apply to other GNSS [global navigation satellite systems] signals as well.

Jamming is a deliberate attack against GPS. It overpowers the signal with a lot of noise and essentially denies operation of a GPS receiver.

Spoofing is a little bit different. It’s a more sophisticated attack, where someone actually fakes a GPS signal and transmits it, and that can provide misleading information. It’s in some ways more dangerous than jamming because you don’t necessarily know that you have a problem. From the receiver’s point of view, you’re receiving a signal, you know where you are, except that it’s misleading information.

On the Extent of the Problem 

Serge Christiaans: [I’ve encountered GPS jamming and spoofing] quite often, actually, especially flying in the eastern part of Europe or flying through Europe diagonally from left to right. We have quite a few areas where there’s jamming and spoofing in a combination. A spoof normally starts with a jam, especially with the L2 band, and then it ends up as a spoof.

Richard Boll: Fortunately, in the United States, where most of my flying has been, we have not experienced the GPS spoofing events that Serge is mentioning. But we are experiencing more GPS jamming events. These are not so much nefarious events, but accidental events. And yes, I have experienced GPS jamming.

Jiri (Yirka) Marousek: Whether getting worse or better, it really doesn’t matter, I think, as long as we address the right problem. In my head, we’re just trying to make it boring. We’re not going to make it go away. There will be jamming, intentional or unintentional. If it’s domestic [U.S.], a lot of it is predictable, and it can actually be briefed for the air op. It doesn’t have to be nefarious.

There are clearly nefarious issues that need to be addressed with procedures and new tech, probably both. But, especially domestically, we can just brief it. A lot of times the data exists, and it’s just not used.

On International Handling

Serge Christiaans: ICAO [the International Civil Aviation Organization] is supposed to be leading in this. But I see them lagging, as any large organization does that creates regulation. They were extremely late in even mentioning cybersecurity in their Annex 17. In one and a half years, there’s one paragraph that says you need to take care of your cyber safety. But they don’t really understand yet how complicated this is.

Most airlines want to be compliant with ICAO. With [Annex] 13 and 17, if it’s not in there, we’re not going to do it. The aviation business is highly stressful with small margins, and we don’t have time for that. 

I still remember the initial GPS jamming on the Iraq-Iran border in September 2023. Nobody had any clue what was happening. Well, we figured it out, and we have procedures now, but we don’t have alternatives yet. I’m happy Ben is working on some great alternatives for that.

We know what's going on, but we still do not handle gracefully anything other than a [single] technical failure. Pilots are very good at taking out a checklist and handling a single technical failure. We quite often do not train for double or triple technical failures. And why don’t we do that? Because statistically, if you calculate the risk, it’s very small.

But now we also have military threat actors who work in the cyber domain who are knowingly and willingly creating chaos. That is something aviation cannot handle yet. The risk formula that we’re using at the moment does not cover that. It covers only things that have happened before.

In the military and in cybersecurity, we use different risk formulas. Aviation looks back at what happened and how can we prevent it. In cybersecurity, cyber defense, and cyber warfare, we look forward. We talk about threat actors, we talk about intent, we talk about capability. We need to change this thinking, and it needs to start on the top, preferably at ICAO. It never works from the bottom.

On Technologies in the Works

Ben Mohr: Honeywell Aerospace has been working on technologies to address these types of issues for 10 years, at least. We have a number of different technologies, both inertially based and then other sensors.

We have the capability now to detect adverse GPS conditions with some of our products and allow the user to just use an inertial-only solution, for example, that can address the problem. Obviously, you don’t have the accuracy you would with the GPS-aided solution.

Longer-term, we have things like using alternative sources of information. We have a vision navigation system, for example, that compares a camera image to a map and can figure out where you are based on that, replacing the GPS input into the navigation system. Similarly, we can use radar systems and other types of sensors to help solve that problem.

The adoption there will be probably first on the defense side. But longer-term, I think those things will make their way into the commercial side.

On Obtaining Information Surrounding Jamming and Spoofing Incidents

Jiri Marousek: That’s where FRATs [flight risk assessment tools] come in—where running the right FRAT up front makes the most sense, whether a safety management system or whatever flight risk assessment tool the operator uses. It should be a point of parity that they all ingest that information and present it to the pilot in a high-priority format.

Richard Boll: That’s part of getting information to the pilot through notams. I’ve read the preliminary [on May 14 accident involving a King Air in New Mexico, where there was a U.S. military jamming exercise]. I don’t want to get into specifics because the NTSB is still investigating, but I have some thoughts on it.

We noticed the dissemination of GPS notams comes through the ARTCC [Air Route Traffic Control Center] system. In other words, they’re not associated with an airport. They’re associated with an air traffic control center. If you’ve ever read the ARTCC notams, you get a whole slew of things. You get restricted areas, windmills, lights that are out of service, MEA [minimum enroute altitude] changes. You can get 30 pages of notams, pilots’ eyes start glossing over when they see them, and they start pushing them to one side.

In the GDAT [GPS Disruption Action Team], we are working to address that, and so is the FAA. The FAA is working on something called a Navigation Operational and Planning Agility Suite, or NOPAS. It’s a tool that’s being developed by Mitre. It’s going to give the FAA Command Center the ability in real time to track unplanned GPS interference events and [other] inadvertent unplanned events where we have GNSS interference that we’ve identified through pilot reports or other information.

There are also plans to make a front-facing tool, web-based, that will allow pilots, like we have the TFR maps here today, in real time, to pull up areas where GNSS interference may be occurring. 

We’re also looking at ways of tying in conventional ground-based navaid outages into that tool there. So, a pilot could quickly look and see if they’re experiencing or may experience GNSS interference and whether ground-based navigation aids may be able to help them overcome it.

Serge Christiaans: Notams don’t work for pilots. Like Rich just said, often you have 20 pages, 30 pages—it’s just a line of frequencies, a line of coordinates. I don’t have time to go through all those. 

Notams are a compliance thing. It’s not an operational thing. It’s not going to work when there’s multiple areas that are jamming or spoofing sensitive. Quite often, it is not being spoofed intentionally, but somebody is trying to protect their critical infrastructure against drones. So, we are collateral damage. 

But we can find out. A Swiss university, the Zurich University of Applied Sciences, built a GPS-wise website that I use every day before every briefing, before every flight, to look where actual jamming and spoofing is going on. 

I’m not so worried about jamming, but I’m worried about spoofing because that cascades down through my Arinc bus and through other systems that I may not be able to predict. That’s where the problem exists. If you look at just a single technical issue, we can solve that. We have procedures for that. But everything that cascades down, that gets dangerous. 

One of the best examples that I keep using is Azerbaijan Airlines that got shot down on a flight from Baku to Grozny on Christmas two years ago. That was a multiple failure in a gray area that nobody saw coming because nobody understood what was happening [and the aircraft] crashed somewhere after being shot by Russians.

About the New Mexico King Air, [jamming] might not be the actual primary reason this accident happened, but it certainly contributed. It could have been prevented, maybe, if the testing of GPS was coordinated better. It’s all about communication and coordination.

In cybersecurity, we call this a threat surface. You have your GNSS; it is a system, it can be jammed, it can be spoofed, and it has serious consequences. It’s a threat surface, and you need to cover it.

Jiri Marousek: Notams are a data source, not the actual solution. In civil aviation, we spend all our time briefing weather, none of our time briefing things like electronic warfare. GPS outages don’t matter. We need to add it to the briefing like we brief weather.

When I have a FedEx feeder operation in California taking off, they run a FRAT and look at a visual depiction of the weather. They need to be able to look at a visual depiction of risk points for interference on a signal and then add [that] into procedures, giving it in a prioritized format. It’s about packaging the information on a timely basis. That cannot solve it all, but it gets us quite far ahead as we get some of the hardware in place to get a redundant source of data or an alternate way of looking at altitude when we know GPS is being spoofed or being interfered with.

Serge Christiaans: Electronic flight bags almost all have options to add a GPS jamming or spoofing layer on your routing. You can have an overlay now on potential turbulence or icing. There’s another tab that shows you the latest data on jamming and spoofing. It’s very helpful.

On Vision-based Technology

Ben Mohr: There are several different ways to do it, but the one that Honeywell Aerospace focuses on has a downward-looking camera. You compare what that camera sees to a database of satellite imagery. You can use that to figure out where you are. That’s typically accurate to around 10 meters, although it can be a little bit better at lower altitudes.

The challenge with vision-based nav is it doesn’t work over the ocean. It doesn’t work in thick clouds. It can work in light cloud cover. If you have a break in the clouds or partial visibility through, it’s pretty good at matching those sorts of things. But if you truly can’t see anything, then it won’t work. You’re not going to find one single solution. There’s no silver bullet here. It’s going to be a combination of different sensors that you have to combine to meet the particular needs of your application. Vision can combine with one of the radio-based approaches, like we talked about Iridium earlier.

We’re also looking into magnetic anomaly navigation. This looks at local variations in the Earth’s magnetic field. It’s caused by things like iron ore deposits in the crust. That actually works pretty well. It’s not as accurate as vision nav, but it works over oceans; it works through clouds. There are complementary technologies like that that you could combine to produce a full solution.

On Backup Technologies Such as Loran Systems

Ben Mohr: Regarding Loran, I always advocate for more signals, and it would be great if we still had that capability.

In terms of other technologies, there are other space-based technologies, low-Earth-orbit satellites. These are essentially private GPS constellations. Iridium provides a signal like that. Some others are coming online that provide signals like that. The premise there is that they’re higher-power, and so they’re harder to jam.

Ultimately, however, they are jammable. I think you’ll find that they’re part of the solution. They decrease the area in which jamming is a problem, but I don’t think there’ll be a whole solution. You’re going to see other technologies, things like celestial-vision nav, radar-based nav approaches. All these things are going to play a part.

This will come on the defense side. The civil side will be fairly far behind because of the certification requirements.

What you’ll see ultimately is hybrid systems that employ multiple different sources of position information. Fuse those, and that provides you with some integrity capability. You can compare those things and figure out what you really trust.

Richard Boll: Alternate position navigation and timing solutions have been something that the PARC and the Navigation Working Group have been closely working on and monitoring over the last 10 years or so. If you go back to the 2010/2011 timeframe, when Ligado proposed putting 5G networks in major metropolitan areas, oh my God, we were going to lose GPS, and we weren’t going to be able to fly our airplanes there anymore. That was kind of the first eye-opening point where we realized that GNSS can be interfered with.

At least domestically—now, we can’t cover the world—we’re looking at alternate position navigation and timing solutions. We’ve instituted the VOR MON [Minimum Operational Network] program to provide the backup VOR in case GPS goes down. We’ve also taken action to increase the DMEs [distance measuring equipment].

We’re not there yet, but we’re looking to get en-route coverage down to 5,000 feet agl, and terminal coverage at the core of 30 airports down to 1,500 feet.

What we are looking down the road is, can we bring that even lower? Can we do RNP approach down to RNP 0.3 accuracy? This would give us basic LNAV and LNAV/VNAV approaches without GPS. That’s where the FAA is starting to look, and there is a lot of energy behind doing that. Is it going to cover every place in the United States? Probably not, but is it going to cover the major airports? The answer hopefully is going to be yes.

On Training and Preparation

Richard Boll: I don’t think we’ve emphasized [GNSS threats] enough in our training. We’ve been emphasizing GPS, at least in the United States. Serge flies in an environment where we have nefarious and protective GPS jamming and spoofing. In the United States, we don’t have that. In the United States, we have basically military-planned GPS jamming events, and they are fairly coordinated. 

There is always direct communication between the ARTCC and the military facility that’s doing the jamming, and there was a phone call that was made, and within six and a half minutes, the GPS jamming was stopped in New Mexico. I agree that GPS interference did play a part in this, because it kind of discombobulated everything, and it led them to try to do a visual approach. I’ll leave it at that.

But the other issue is when we get GPS jamming events that we cannot trace. For about three hours in 2022 in the Denver area, we had a GPS jamming event that we couldn’t figure out. It happened the day after we rolled out the big 5G radar altimeter interference deal. Everybody thought that was tied to that. It wasn’t until the Denver Metro found out that their wheel brake monitoring system wasn’t reporting on the rail cars that we [realized] the only input [for that] was GPS. We figured: “Hey, this is GPS interference.”

A month later, for about 48 hours, we had intermittent, widespread GPS jamming in Dallas. We don’t really have a real explanation of what caused that. We have no way to communicate that information to pilots through notams, other than the ARTCC. 

This new NOPAS tool hopefully will provide pilots real-time information, probably initially on an FAA website, but we are pushing to make this an open source where we can incorporate it directly into EFBs, and you can see where the big picture radar is. One of the things working in the United States is for better communications between the air carriers and the air carrier operations center and the Air Traffic Control System Command Center, so that we pick up faster, quicker potential GPS jamming events.

One of the things we pick up on GPS jamming is what happens to ADS-B Out. In the Dallas event, that was the first indication [of jamming] because airplanes that were supposed to be in the DFW terminal area were showing up over Oklahoma. It’s that type of information we want to collect and feed into the Air Traffic Control System Command Center. I can tell you there is a Men in Black organization that, in the U.S., monitors this stuff very closely. If they start detecting any type of GNSS interference, they are deploying to the location to try to find it.

Jiri Marousek: We touched on the first line of defense, better procedures and training with the information we have. The information we have is not used as well as it should. It’s not briefed as well as it should, and it’s just not brought into the cockpit in a way that’s actionable for the pilot. That will address a lot of the at least non-nefarious issues.

When it comes to nefarious issues, then we shouldn’t rely on a single source of data. We all know that, and we are. When it comes to GPS, there’s longer-range hardware that Ben and teams like his are going to have to deliver. But what's in the cockpit now, or what’s in the hardware now, that can start becoming a decision versus reliance [and let] the crew know it’s happening, they have a misaligned piece of data, and they need to make a decision on what they trust?

That’ll get better over time. There are companies like Cyviation and others that are doing the in-flight cybersecurity on top of monitoring for soundness of signals. Software is going to be ahead of hardware because certification and retrofitting are not cheap and take time.

EFBs are probably the place where that information is going to have to come in, where the algorithms—call it AI—need to better inform the air crew because they can’t have too much information.

As an industry, training and preparation are the things that we need to do. It doesn’t have to be regulatory. In private aviation, Part 135, 91, et cetera, it might be driven by insurance, [based on] liability and risk, which is a lot quicker to respond than regulations.

On Airline Action

Serge Christiaans: [Before airlines retrofit technologies], they should be available on the market first. I guess that’s what we’re waiting for. Ben is talking about some beautiful, promising projects. There’s a lot coming.

But it’s not there yet. And until then, we’re compliant. And that’s it. Airplanes have a long life cycle. A lot of airplanes fly around 20 years old, 25 years old. They were built before the words cybersecurity, cyberwarfare, cyber defense, or electronic warfare existed.

We knew that the GNSS was a big threat surface, but there was nobody abusing it. But now it’s a single point of failure. So, we need to do defense in layers. And that means, like Ben says, diversification.

Ben Mohr: Some of these products do exist. The challenge on the civil side is regulatory. You can buy a vision nav system today. You can use it on a UAV. They’re common in Ukraine.

But you can’t buy one that has civil certification. There are no [minimum operational performance standards]. That’s an area of challenge for the adoption on the civil side.

Serge Christiaans: Then the airplanes are flying with old Arinc 429 buses, and it all has to fit. It all has to talk to each other, digitally fit. That is more challenging than it looks.

Ben Mohr: If you were to describe the adoption curve, the first place you’ll see it, where you’ve already seen it, is on the defense side, and specifically the UAVs. Then you’ll slowly see it move up to defense aircraft, crewed aircraft. And then I think you’ll start to see it on the civil side. Probably, commercial aviation is going to be the last. You’ll see the general aviation and other business aviation applications pick those things up sooner.

On Use of Quantum

Ben Mohr: At Honeywell Aerospace, we’re looking at two different quantum navigation approaches. We define quantum navigation as navigation that’s using quantum sensors.

There are two major ones. Magnetic anomaly navigation does use quantum-based magnetometers. That’s what we fly in our development flights. Those are not science fiction. You can buy them today. They're not cheap, but you can order one. 

There are also, a little bit farther behind, quantum inertial sensors. These are similar to gyros and accelerometers, similar to what is in a traditional INS [inertial navigation system]—for example, our ADIRU [air data inertial reference unit] or Laseref products. Except, they operate on different principles, on quantum principles, that allow them to be much more accurate. Those hold a lot of promise for operating in these types of environments.

One of the immediate things a pilot can do now is just switch from their hybrid GPS INS solution to their INS-only solution. What will happen in the future when we have quantum-based inertial navigation systems is that the INS-only solution will be much more accurate. The difference between the hybrid and the INS-only will be smaller.

On the magnetometer side, quantum is actually quite small already and feasible to deploy on aircraft. Quantum inertial sensors are still large and not suitable for aircraft use yet. Although Honeywell Aerospace’s done some test flights with those types of sensors. They’re coming, but we’re a long way out.

There’s a lot of government investment going into that area, so that helps push that technology forward. 

On Pilot Advice

Richard Boll: [NBAA doesn’t] publish directly any advice. We refer operators to existing guidance. First and foremost, if you have any OEM guidance, such as aircraft flight manual supplements, or operational guidance in the FCOM [flight crew operating manual], that should be your primary source for anything on GNSS interference.

The FAA just published its GNSS interference guide. If you do not have anything from your aircraft manufacturer—for example, you are flying a lighter GA airplane or your company doesn’t have robust procedures like you would see in a 121 carrier—this guide will help you develop the SOPs [standard operating procedures] or develop the background information on the differences between GPS jamming or spoofing.

We would like operators to download this. It’s on the [FAA] Flight Standards website: FAA GPS Interference Guide. We’re working with the FAA to keep that updated.

Jiri Marousek: This needs to be part of the nomenclature of primary training. [Also], you can turn on your EFB as a layer. For all of us that provide safety and security data into the cockpit during training, prior to flight, briefing the flight, we’re part of the solution.

Many providers present information to crews. And this has to be part of the nomenclature, part of the product design, and part of how we present the information. It has to be normalized just like weather.

Richard Boll: It’s also important that we, as pilots, prepare ourselves for operating in a non-GPS environment. For a lot of airplanes, like what Serge flies, you can revert to an IRU [inertial reference unit]. 

But there are a lot of airplanes out there [that] if GPS goes away, your RNAV system is gone. They don’t have a DME/DME backup. I’m seeing way too many pilots starting to disregard VOR navigation. They think, “I don’t need to have to fly a VOR approach. I don’t need to know how to do a conventional ILS to a missed approach because I’ve always got the missed approach in the FMS.” When was the last time you went to a 141 or 142 school, or training center, flew an ILS approach with a conventional missed [approach]? 

The one time that you might have to do that may be the one time the GPS is not available, and you might have to go 100 miles from your planned destination airport that only had RNAV GPS approaches.

Serge Christiaans: I agree. It’s skills that we lost because we rely on the cheap and accurate GNSS. 

It’s about training and keeping your skills up. We have airports in Finland in remote areas that only have RNP approaches or LNAV approaches. Once the Russians start jamming because they have a 1,300-km border with these guys, you have to close the airport. There are no alternatives. Communities are being shut off, isolated.

[It is easy to revert to VOR navigation] but your GPS primary is gone, and you will lose accuracy, which is fine when you’re en route, when you have 10 miles left and right. But, once you start descending into the terminal or into an approach, that might be a problem, and you need to be aware of that.

We’re just talking about jamming, which is actually an easy scenario. Spoofing is where the real [problems] start. It’s where the cascading down happens. What is it doing with the rest of the hardware in my airplane? 

[Regarding] the Azerbaijan accident, these guys did not know that there was spoofing, and they lost their ADS-B. The Russian air defense was on the ground defending the airport and did not see an identification friend or foe on this airplane. So, they started shooting. The pilots knew the GPS was [messed] up, but they had no clue that it was spoofing and that the ADS-B wasn’t working. Still, they were dispatched into a city heavily protected by AAA [anti-aircraft artillery]. It was a perfect storm.

They knew it was going on. This is Azerbaijan. They had permission for that flight. But, the city was protected, which means there is jamming going on, and in this case, spoofing as well. The anti-aircraft battery guys on the ground simply had no positive identification and started shooting. 

In cybersecurity, we talk about technology, people, and process. The process going back, these guys should never have been dispatched in the first place. But this is where commercial pressure comes in and also diplomatic pressure: You cannot stop flying between our two countries because then somebody is not happy. It’s quite complex.

Richard Boll: The accident that Serge is talking about has been a topic of great discussion in the GDAT. We talked about [how] the crew tried to communicate that they had lost their GPS. But the controllers were still trying to clear them for an RNP approach—basically an RNAV GPS approach. They didn’t understand that [they] didn’t have the equipment necessary to fly that approach. So, there’s a lot of moving parts here that need to be addressed.

Serge Christiaans: They made two tries. They made it go around in the first. And there was also weather involved, so this crew was under heavy strain. And then they got shot. The airplane was still controllable, but only half, and they limped away outside of Russian airspace. At the end, they couldn’t land it normally. So, he made a crash landing, and people died, but people survived as well, which is a miracle.

On Future Technologies

Ben Mohr: We talked about vision, magnetic anomaly, [and] low-Earth-orbit satellites. Celestial is another one on the horizon, using stars [for position information]. You can now also use satellites, space debris, anything that reflects enough light to see with your telescope. It can give you increased accuracy. So that’s an area of work for us.

There are others as well. You can model the vehicle dynamics, use that as a constraint on your inertial [solution]. That’s probably for lower-end platforms, not so much for crewed aircraft.

But there are many different techniques, radar-based techniques. As I said before, I don’t think there’s really any one silver bullet. I think what you’re going to see is a hybrid system that combines multiple sources.