What is GPS spoofing and how do drones detect it?
Published:
GPS spoofing is when a GNSS receiver is fed fake signals that imitate real satellite signals, so that it computes the wrong position or time without noticing. Drone software does not detect this with a single check; it tries to catch it by watching signal quality, the consistency of the navigation solution and agreement with other sensors together.
The difference between spoofing and jamming
Jamming drowns out satellite signals with strong noise; the receiver loses its position and usually notices right away. Spoofing is more subtle: the receiver keeps working, but the position it computes is wrong.
That is why the real danger of spoofing is that it goes unnoticed. An autopilot that uses position data without questioning it can keep making decisions based on the wrong position.
Why is spoofing possible?
The structure of civilian GPS signals is public, and historically these signals have carried no authentication. Signals from the satellites are also very weak by the time they reach the ground. Together, these two properties make it hard for a receiver to tell a real signal from an imitation.
To address this, signal authentication has started to be added to satellite systems; for example, OSNMA (Open Service Navigation Message Authentication) was developed for Galileo's open service. Such protections require receiver support, though, and do not stop every type of attack on their own.
How does software spot a fake signal?
No single method catches every case, so detection is usually layered:
- Signal-level indicators: if the receiver provides them, signs such as an unusually high carrier-to-noise ratio (C/N0), suspiciously similar values across all satellites, or sudden changes in automatic gain control (AGC) are monitored.
- Solution consistency: sudden jumps in position or clock, speeds and accelerations the vehicle physically cannot achieve, and integrity tests (RAIM-like) that look for inconsistencies between satellite measurements are used.
- Cross-checking with sensors: the position and velocity from GNSS are compared with on-board sources such as the IMU, barometric altimeter, airspeed sensor, magnetometer and visual odometry. Sensor fusion filters can reject measurements that do not match expectations using innovation tests.
- Multiple constellations and frequencies: the consistency of solutions from different systems, such as GPS, Galileo, GLONASS and BeiDou, or from different frequencies can be checked against each other.
- Hardware-assisted methods: multi-antenna arrays can look at the direction signals arrive from; real satellite signals come from different parts of the sky, while fake signals from a single transmitter tend to come from the same direction. This method needs extra hardware.
What happens after detection?
When spoofing is suspected, the software's goal is to stop using the wrong position. Typical responses are removing GNSS measurements from sensor fusion, switching to GPS-denied navigation sources, alerting the operator and, if needed, starting a predefined safe behavior (failsafe).
For these steps to work, the vehicle needs a position estimate that is independent of GNSS. An on-board navigation source that does not rely on satellites makes it easier to notice the inconsistency with the fake signal and to keep flying afterwards.
Navigation without satellites: IRA-2 →