1. INTRODUCTION
In recent years, unmanned aerial vehicles (UAVs) have been widely used which brought a serious threat to the safety of the low-altitude domain [1], [2]. At present, to improve the data transmission rate and anti-interference ability of UAVs, wireless communication links have been designed with characteristics such as broadband, multiple links, and hopping spread spectrum [3]. However, during UAV signal detection, strong background noise and co-channel interference often cause interference [4], [5]. Advanced deep learning techniques offer a promising solution to these issues. The dataset’s quality can significantly affect the accuracy of results inference by deep learning algorithms [6]. Moreover, the current UAV signal datasets DroneRF and VTI_DroneSET_FFT cannot be applied directly to target detection [7], [8]. Therefore, finding an automatic and efficient method for labeling datasets is a key challenge.