Evaluation and telemetry-based detection of GPS spoofing effects on UAV navigation using software-defined radio

Area of Science:

• Robotics and Autonomous Systems
• Cybersecurity
• Navigation Systems

Background:

• Autonomous drones heavily rely on Global Positioning System (GPS) for navigation.
• GPS spoofing attacks pose a significant threat to drone safety and mission integrity.
• Existing detection methods may be costly or platform-specific.

Purpose of the Study:

• To experimentally investigate the vulnerabilities of autonomous drone navigation under GPS spoofing.
• To develop and evaluate a low-cost, platform-independent method for detecting GPS spoofing attacks.
• To enhance drone resilience against GPS spoofing threats through sensor fusion analytics.

Main Methods:

• Implemented a low-cost GPS spoofing framework using Software Defined Radio (SDR), HackRF One, and GPS-SDR-SIM.
• Conducted static and dynamic spoofing trials on an F450 quadcopter with Pixhawk 4 and Ublox NEO-M8N GPS.
• Developed a telemetry-based detection indicator by analyzing deviations between GPS, Inertial Measurement Unit (IMU), and barometric sensor data.

Main Results:

• Spoofed GPS conditions caused significant deviations in drone trajectory, geofence breaches, and unstable behavior, leading to a crash.
• Analysis revealed anomalies in barometric altitude, yaw angle, magnetometer, and GPS indicators (satellite count, HDOP) during spoofed flights.
• The developed detection indicator achieved over 90% reliability in identifying spoofing-induced anomalies.

Conclusions:

• Autonomous drone navigation systems are susceptible to GPS spoofing, highlighting critical security vulnerabilities.
• Cross-sensor anomaly detection using existing telemetry data offers an effective, low-cost, and platform-independent solution for real-time spoofing detection.
• This research contributes a lightweight methodology for improving drone resilience against GPS spoofing via sensor fusion analytics.