Jullian, OliviaOtero, BeatrizStojilović, MirjanaCosta, Juan JoséVerdú, JavierPajuelo, Manuel Alejandro2022-02-082022-02-082022-02-082022-02-0210.1007/978-3-030-95467-3_38https://infoscience.epfl.ch/handle/20.500.14299/185213Unmanned aerial vehicles (UAVs) are widely deployed in air navigation, where numerous applications use them for safety-of-life and positioning, navigation, and timing tasks. Consequently, GPS spoofing attacks are more and more frequent. The aim of this work is to enhance GPS systems of UAVs, by providing the ability of detecting and preventing spoofing attacks. The proposed solution is based on a multilayer perceptron neural network, which processes the flight parameters and the GPS signals to generate alarms signaling GPS spoofing attacks. The obtained accuracy lies between 83.23% for TEXBAT dataset and 99.93% for MAVLINK dataset.Deep learningIntrusion detection modelUnmanned aerial vehiclesSpoofingGlobal navigation satellite systemtext::conference output::conference proceedings::conference paper