Rapid Detection of UAV Controllers and Operators Based on Remote Control Signal Direction Finding

With the increasing security threats posed by unmanned aerial vehicles (UAVs), traditional anti-UAV technologies are facing challenges such as high economic costs, inefficiency, and difficulty in addressing the root causes of the issue. To address these issues, this paper proposes a rapid detection technology for non-cooperative UAV controllers and operators by combining signal direction finding with visual detection. Firstly, the method determines the direction of remote control signals from non-cooperative UAVs on airborne platforms using signal direction finding technology based on time-modulated array antennas, thereby overcoming the limitations of hardware complexity and detection accuracy. Furthermore, a lightweight detection method is developed by optimizing the YOLOv12 architecture, simplifying feature fusion modules, and establishing a network selection mechanism based on perspective projection theory. This improves detection speed under resource-constrained conditions. Experimental results demonstrate that the improved model achieves a 21.4-57.4% increase in detection frame rate, a 37.4-76.4% reduction in parameters, and a 15.5-39.7% decrease in computational load, while maintaining acceptable accuracy. Field tests on the DJI M350 UAV confirm a 39.46% improvement in frame rate, validating the technology's practical effectiveness. The proposed method presents a novel perspective for anti-UAV measures.