Drone RF Signal Detection and Fingerprinting: UAVSig Dataset and Deep Learning Approach

Unmanned aerial vehicles (UAVs) are useful for commercial, recreational, and military applications, but they can also be used by malicious attackers and pose security threats. Therefore, it is important to monitor their occurrences and ensure their compliance. Radio frequency signals can be leveraged for such task. However, commercial UAVs often adopt the frequency hopping spread spectrum signals and proprietary protocols, which makes them more challenging to detect. While prior works have studied the UAV detection and classification problem, most of them consider the classification between different UAV models. Classification between drones of the same model has not yet fully been investigated. In this work, we consider the tasks of detecting and localizing UAV signals in a wideband spectrum, and also the task of fingerprinting these drones of the same model simultaneously. To solve this problem, we collect the UAVSig, an over-the-air dataset of UAV RF signals. We also present a deep learning model which can solve detect and fingerprint multiple drones of the same model simultaneously based on spectrograms. Our model can achieve 99.5% precision and 99.4% recall for transmission detection, and 90.9% classification accuracy with drones of the same model.