Satellite navigation interference tends to be discussed as a single problem, usually under the label of GPS denial. In practice there are two quite different threats behind that phrase, and they fail in opposite ways. Jamming removes the signal. Spoofing replaces it. For a person navigating with a phone the difference is academic. For an autopilot flying a fixed-wing aircraft beyond visual line of sight, it is the difference between a problem the aircraft knows about and one it does not. This article explains both in general terms, why spoofing is the more dangerous of the two, and what operators and system designers can do about each.
Why satellite signals are easy to disrupt
Global navigation satellite systems (GNSS), including GPS, Galileo, GLONASS and BeiDou, broadcast mostly from medium Earth orbit, roughly 20,000 kilometres up. By the time the signals reach the ground they are extremely weak, below the level of background radio noise, and a receiver can only recover them because it knows precisely what pattern to look for.
That weakness is the root of both problems. Only a small amount of transmitted power, reasonably close to the receiver, is needed to overwhelm the genuine signal. And for the open civil signals that most drones rely on, the structure of the signal is public, so anyone with suitable equipment can generate something that looks like it.
Jamming: the signal disappears
A jammer transmits noise or an interfering signal on the GNSS frequencies. The receiver’s view of the satellites is drowned out, the measured signal quality falls, the number of satellites tracked drops, and eventually the receiver reports that it has no fix.
Jamming is not only a military concern. Interference can also come from faulty or poorly shielded electronics and other everyday sources, so an aircraft can meet it well away from any conflict zone.
From the autopilot’s point of view, jamming is a loud failure. The receiver reports a degraded or lost fix, the status fields say so, and the failsafe logic has something concrete to act on. The aircraft may not know where it is, but it knows that it does not know. That is a bad situation, but a well-understood one, and most autopilots have a configured response to it.
Spoofing: the signal is replaced
A spoofer transmits counterfeit satellite signals, strong enough that the receiver tracks them instead of the genuine ones. Done crudely, the reported position jumps or the receiver’s clock does something implausible. Done carefully, the spoofer first matches the real signals closely, takes over the receiver’s tracking, and then shifts the reported position gradually away from the truth.
A related technique, known as meaconing, records genuine signals and rebroadcasts them with a delay. It produces a plausible but wrong position without the attacker needing to generate signals from scratch.
From the receiver’s side, a well-executed spoof can look like a perfectly healthy fix: plenty of satellites, good signal strength and a sensible accuracy estimate. Nothing in the status fields needs to change.
Why spoofing is worse for an autopilot
Consider how a typical autopilot uses GNSS. Its navigation filter blends satellite position and velocity with inertial measurements, weighted by how trustworthy each appears to be. It rejects measurements that disagree sharply with its current estimate. And it steers the aircraft to reduce the difference between where it believes it is and where the flight plan says it should be.
Now feed that system a slowly drifting false position:
- The filter accepts each update, because each differs only slightly from the last and from the inertial prediction. A sudden jump would be rejected; a gradual drag usually is not.
- The autopilot sees an apparent cross-track error and corrects for it, physically flying the aircraft away from its intended path so that the false position matches the plan.
- Geofences, return-to-home points and landing approaches are all evaluated against the false position, so the protections the operator relies on move with the deception.
- Telemetry and logs show the aircraft on track, so the remote pilot sees nothing wrong at the ground control station.
That is the core of the problem. Jamming produces an aircraft that knows it is lost. Spoofing produces an aircraft that is confident, wrong and actively steering towards the wrong place. Every safety mechanism downstream of the navigation estimate inherits the error.
Detecting spoofing
No single test catches every spoof, but a combination of checks raises the bar considerably:
- Signal-level monitoring. Signal strengths that are unusually uniform across satellites or suddenly higher than normal, or abrupt changes in the receiver’s automatic gain control, can indicate a spoofer. Some receivers include this monitoring.
- Multi-constellation and multi-frequency consistency. A spoofer has to fake every signal the receiver tracks. Receivers that use several constellations and frequency bands and cross-check them are harder to deceive, although not immune.
- Inertial consistency. Comparing GNSS against the aircraft’s inertial measurements catches abrupt manipulation. It is much weaker against a slow drag, because inertial errors also grow over time and the two drifts can hide inside each other.
- An independent absolute reference. The strongest check is a second source of absolute position that does not use radio signals at all. If it disagrees persistently with GNSS, something is wrong, and because the second source is not exposed to radio spoofing, the disagreement points clearly at the satellite solution.
Designing the response
Detection is only useful if the aircraft does something sensible with it. For jamming the response is usually straightforward to define: the fix is gone, so the aircraft falls back to another source or a failsafe. For spoofing, the operator needs to decide in advance:
- How much disagreement between sources, sustained for how long, counts as a spoofing alarm.
- Which source the aircraft trusts once an alarm is raised, and whether GNSS is ignored for the rest of the flight or only until the sources agree again.
- What the remote pilot sees, because an alarm that is not surfaced clearly will not be acted on.
- What the aircraft does if it cannot resolve which source is right.
It is also worth treating the return of GNSS with some suspicion. A signal that reappears after a period of jamming may be the genuine signal coming back, or it may be a spoofer that jammed first to break the receiver’s lock. Requiring a period of agreement with an independent source before GNSS is trusted again is cheap insurance.
Where TerrainSLAM fits
TerrainSLAM determines absolute position by matching what a downward camera sees against pre-loaded terrain data, entirely onboard. It uses no satellite signals, radio beacons or ground infrastructure, so there is nothing for a jammer to drown out or a spoofer to imitate. That makes it useful in both of the roles described above: a navigation source when GNSS is jammed, and an independent reference that can expose a spoofed fix. It is built for fixed-wing aircraft and integrates with leading autopilot platforms without firmware changes. If GNSS integrity is part of your risk assessment, we are happy to talk through how it would sit alongside your existing setup.