To overcome AMR's limited generalization and low classification accuracy in non-cooperative communication contexts with low signal-to-noise ratio, This paper proposed a model named Convolutional Gated Recurrent Units Deep Neural Networks (CGDNN) . The model integrated convolutional neural networks (CNN) , gated recurrent units (GRU) , and deep neural networks. To mitigate noise impact on modulation detection, this paper initially denoised I/Q sampled signal using wavelet thresholding. Subsequently, this paper utilized CNN and GRU for extracting spatial and temporal features from signals before proceeding to classification through fully connected layers. Besides enhancing AMR performance, the CGDNN model significantly reduced computational complexity compared to competitors. Experiment results demonstrated an average recognition accuracy of 64.32% on the RML2016.10b dataset, with an enhancement in signal classification accuracy from -12dB to 0dB. Moreover, the model substantially decreased confusion between 16QAM and 64QAM, achieving a peak recognition accuracy of 93.9% at 18dB. CGDNN model effectively improved AMR detection accuracy under low signal-to-noise ratio conditions and enhanced model training efficiency.

[2024-02-06] Accepted Paper /doc/2023.11.0581/v1