Uav Audio Signal Detection Method Based on Gru and Attention Mechanism

The rapid development of drone technology has resulted in its ever-expanding applications in the military, commercial, and civilian domains. Nevertheless, the concomitant safety and privacy issues have become increasingly conspicuous, giving rise to an urgent demand for effective monitoring and management of drone activities. Consequently, anti-drone detection and recognition technology has gradually emerged as a research focus. Deep learning provides innovative solutions for drone detection by virtue of its advantages in complex data feature extraction and intelligent signal analysis. In this paper, a UAV audio signal detection method based on GRU neural network and attention mechanism is proposed to solve the limitations of traditional convolutional neural network (CNN) in dealing with background noise and interference signals. In this study, one-dimensional waveform data is converted into a two-dimensional feature space through spectrogram feature extraction of audio signals, so as to capture the global temporal characteristics and dynamic features of UAV audio signals. In the model design, a lightweight two-layer GRU neural network is adopted, which combines the update gate and reset gate mechanisms to effectively capture the short-term and long-term dependencies in the audio signal, avoid the gradient vanishing problem, and improve the computational efficiency and generalization capability. In addition, to further enhance the ability of the model to focus on long time series and complex background signals, an attention mechanism is introduced so that the model can dynamically focus on the information of key time steps, thus improving the robustness of complex signal processing. In order to verify the effectiveness of the proposed method, we conducted experiments on publicly available UAV audio datasets, and the results show that the method proposed in this paper exhibits high detection accuracy and robustness in dealing with the task of UAV audio detection in complex environments, which provides reliable technical support for the development of UAV detection technology.