2026 Vol.9 Feb.N01   

• Title: Research on Automatic Signal Modulation Recognition Methods Based on Deep Learning
• Name: WenLu Zhang1* ,Yan Zhou1, Yun An2
• Company: 1.Faculty of Engineering, Sias University ，ZhengZhou 451100 China 2. China Mobile IoT Corporation Limited, Chongqing 400060 China
• Abstract:

  In the context of automatic modulation recognition (AMR) tasks conducted under complex electromagnetic environments, traditional manual feature extraction methods demonstrate inadequate robustness under conditions of low signal-to-noise ratio (SNR). Moreover, existing deep learning models are constrained by limitations in feature representation and temporal modeling capabilities. To address these challenges, this paper introduces a hybrid deep learning model that integrates multi-scale convolution, an attention mechanism, and bidirectional gated recurrent units (BiGRUs). The proposed model employs a multi-scale convolutional architecture to capture local features across varying temporal receptive fields, incorporates a channel-temporal joint attention mechanism to adaptively emphasize critical discriminative information, and utilizes BiGRUs to model bidirectional temporal dependencies within signals, thereby enhancing the representational capacity for complex modulated signals. Experimental evaluations conducted on the publicly available dataset RML2016.10a reveal that the proposed model maintains consistent stability across the entire SNR spectrum, achieving a recognition rate exceeding 90% under SNR ≥ 0 dB conditions and demonstrating an average performance improvement of approximately 2% to 5% relative to comparative methods within the moderate SNR range. Furthermore, this approach effectively balances computational efficiency with recognition performance, underscoring its potential for practical engineering applications.

  
  

• Keyword: Deep Learning; Attention Mechanism; Gated Recurrent Unit (GRU) Network; Convolutional Neural Network (CNN)
• DOI: 10.12250/jpciams2026090213
• Citation form: WenLu Zhang ,Yan Zhou, Yun An.Research on Automatic Signal Modulation Recognition Methods Based on Deep Learning[J]. Computer Informatization and Mechanical System,2026,Vol.9,pp.

[1] ZHANG F, LUO C, XU J, et al. Deep learning based automatic modulation recognition: Models, datasets, and challenges[J]. Digital Signal Processing, 2022, 129: 103650. DOI:10.1016/j.dsp.2022.103650.

[2] JDID B, HASSAN K, DAYOUB I, et al. Machine learning based automatic modulation recognition for wireless communications: A comprehensive survey[J]. IEEE Access, 2021, 9: 57851-57873. DOI:10.1109/ACCESS.2021.3071801.

[3] HUYNH-THE T, PHAM Q V, NGUYEN T V, et al. Automatic modulation classification: A deep architecture survey[J]. IEEE Access, 2021, 9: 142950-142971. DOI:10.1109/ACCESS.2021.3120419. 

[4] O’SHEA T J, CORGAN J, CLANCY T C. Convolutional radio modulation recognition networks[C]//Engineering Applications of Neural Networks. Cham: Springer, 2016: 213-226. DOI:10.1007/978-3-319-44188-7_16.

[5] LIU X, YANG D, EL GAMAL A. Deep neural network architectures for modulation classification[C]//2017 51st Asilomar Conference on Signals, Systems, and Computers. Pacific Grove: IEEE, 2017: 915-919. DOI:10.1109/ACSSC.2017. 8335483.

[6] SUN S, WANG Y. A novel deep learning automatic modulation classifier with fusion of multichannel information using GRU[J]. EURASIP Journal on Wireless Communications and Networking, 2023, 2023: 66. DOI:10.1186/s13638-023-02275-y.

[7] WANG Z, LIU H, ZHANG Y, et al. Robust automatic modulation classification via TCN-GRU hybrid network[J]. Sensors, 2024, 24(24): 7908. DOI:10.3390/ s24247908.

[8] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 770-778. DOI:10.1109/CVPR. 2016.90.

[9] O’SHEA T J, ROY T, CLANCY T C. Over-the-air deep learning based radio signal classification[J]. IEEE Journal of Selected Topics in Signal Processing, 2018, 12(1): 168-179. DOI:10.1109/JSTSP.2018.2797022.

[10] ZHANG R, YIN Z, WU Z, et al. A novel automatic modulation classification method using attention mechanism and hybrid parallel neural network[J]. Applied Sciences, 2021, 11(3): 1327. DOI:10.3390/app11031327.

[11] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]// Advances in Neural Information Processing Systems 30. Red Hook: Curran Associates, 2017: 5998-6008.

[12] CAI J, GAN F, CAO X, et al. Signal modulation classification based on the transformer network[J]. IEEE Transactions on Cognitive Communications and Networking, 2022, 8(3): 1348-1357. DOI:10.1109/TCCN.2022.3176640. 

[13] XU Y. AMC-Transformer: Automatic modulation classification based on enhanced attention model[J]. Infocommunications Journal, 2025, 17(4): 32-40. DOI: 10.36244/ICJ.2025.4.5.

[14] WANG D, LIN M, ZHANG X, et al. Automatic modulation classification based on CNN-Transformer graph neural network[J]. Sensors, 2023, 23(16): 7281. DOI:10.3390/s23167281.

[15] ZHANG X, CHEN X, WANG Y, et al. Lightweight automatic modulation classification via progressive differentiable architecture search[J]. IEEE Transactions on Cognitive Communications and Networking, 2023, 9(6): 1519-1530. DOI:10.1109/TCCN.2023.3306391.

[16] CHANG S, HUANG S, ZHANG R, et al. Multi-task learning based deep neural network for automatic modulation classification[J]. IEEE Internet of Things Journal, 2022, 9(3): 2192-2206. DOI:10.1109/JIOT.2021.3091523

[17] SUN S, WANG Y. A novel deep learning automatic modulation classifier with fusion of multichannel information using GRU[J]. EURASIP Journal on Wireless Communications and Networking, 2023, 2023: 66. DOI:10.1186/s13638-023-02275-y.

[18] HUYNH-THE T, HUA C-H, PHAM Q-V, et al. MCNet: An efficient CNN architecture for robust automatic modulation classification[J]. IEEE Communications Letters, 2020, 24(4): 811-815. DOI:10.1109/LCOMM.2020. 2968030.