Automatic Modulation Classification using Residual Connection and Bottle-neck Transformers with Trend-aware Self-Attention

Geng Wang; Luming Li; Xin Chen

doi:10.62177/jaet.v3i2.1262

Authors

Geng Wang Beijing Institute of Satellite Information Engineering; State Key Laboratory of Space Information System and Integrated Application
Luming Li Beijing Institute of Spacecraft System Engineering
Xin Chen Beijing Institute of Satellite Information Engineering; State Key Laboratory of Space Information System and Integrated Application

DOI:

https://doi.org/10.62177/jaet.v3i2.1262

Keywords:

Automatic Modulation Classification, Deep Learning; Depthwise Separable Convolution, Bottleneck Transformers

Abstract

With the increase of modern communication equipment, the demand for AMC (Automatic Modulation Classification) in communication systems is increasing. Since deep learning was introduced into AMC, people have been working on improving the recognition accuracy and robustness of AMC. To achieve this, the study proposes the design of a ResDBoTTA (Residual DSC Bottleneck Transformers with Trend-aware Attention) AMC Network for communication signal modulation pattern classification. The DSC (Depthwise Separable Convolution) used in the model can significantly reduce the model parameters. The introduced time trend-aware self-attention mechanism can eliminate the influence of abnormal noise. Finally, global deep convolution is applied to enhance recognition accuracy. To verify the classification performance of ResDBoTTA AMC Net, relevant simulation studies are carried out. Experimental results demonstrate that ResDBoTTA AMC Net achieves superior recognition performance compared to existing technologies.

Downloads

Download data is not yet available.

References

Mohsen, S., Ali, A. M., & Emam, A. (2023). Automatic modulation recognition using CNN deep learning models. Multimedia Tools and Applications, 83(3), 7035–7056. https://doi.org/10.1007/s11042-023-15814-y

Das, D., Bora, P. K., & Bhattacharjee, R. (2021). Automatic modulation classification over MIMO amplify and forward (AF)-relay fading channels. Physical Communication, 47, 101399. https://doi.org/10.1016/j.phycom.2021.101399

Chen, Q., Meng, W., Han, S., Li, C., & Chen, H.-H. (2022). Robust task scheduling for delay-aware IoT applications in civil aircraft-augmented SAGIN. IEEE Transactions on Communications, 70(8), 5368–5385. https://doi.org/10.1109/tcomm.2022.3186997

Moulay, H., Djebb ar, A. B., Dehri, B., & Besseghier, M. (2024). Dendrogram-based heterogeneous learners for automatic modulation classification in DSTBC-OFDM systems. Physical Communication, 62, 102241. https://doi.org/10.1016/j.phycom.2023.102241

Li, J., Chen, Q., Long, Z., Wang, W., Zhu, H., & Wang, L. (2021). Spectrum sensing with non-Gaussian noise over multi-path fading channels towards smart cities with IoT. IEEE Access, 9, 11194–11202. https://doi.org/10.1109/access.2021.3051719

Dobre, O. A., Abdi, A., Bar-Ness, Y., & Su, W. (2007). Survey of automatic modulation classification techniques: Classical approaches and new trends. IET Communications, 1(2), 137. https://doi.org/10.1049/iet-com:20050176Wang,

D., Zhang, M., Li, J., Li, Z., Li, J., Song, C., & Chen, X. (2017). Intelligent constellation diagram analyzer using convolutional neural network-based deep learning. Optics Express, 25(15), 17150. https://doi.org/10.1364/oe.25.017150

Hassan, K., Dayoub, I., Hamouda, W., & Berbineau, M. (2010). Automatic modulation recognition using wavelet transform and neural networks in wireless systems. EURASIP Journal on Advances in Signal Processing, 2010(1). https://doi.org/10.1155/2010/532898

Parmar, A., Chouhan, A., Captain, K., & Patel, J. (2024). Deep multilevel architecture for automatic modulation classification. Physical Communication, 64, 102361. https://doi.org/10.1016/j.phycom.2024.102361

O’Shea, T. J., Corgan, J., & Clancy, T. C. (2016). Convolutional radio modulation recognition networks. In Engineering Applications of Neural Networks (pp. 213–226). https://doi.org/10.1007/978-3-319-44188-7_16

Liu, X., Yang, D., & El Gamal, A. (2017). Deep neural network architectures for modulation classification. In 2017 51st Asilomar Conference on Signals, Systems, and Computers (pp. 1–5). https://doi.org/10.1109/acssc.2017.8335483

Zhou, F., Li, J., & Wang, Y. (2023). An improved CNN‐LSTM network for modulation identification relying on periodic features of signal. IET Communications, 17(18), 2097–2106. https://doi.org/10.1049/cmu2.12682

Hamza, M. A., Alghamdi, A. M., Alzahrani, J. S., Alharbi, M. T., & Al‐Turki, Y. (2022). Optimal bidirectional LSTM for modulation signal classification in communication systems. Computers, Materials & Continua, 72(2), 3055–3071. https://doi.org/10.32604/cmc.2022.024490

Zhu, Z., Sun, D., Gong, K., Wang, W., & Sun, P. (2021). A lightweight CNN architecture for automatic modulation classification. Electronics, 10(21), 2679. https://doi.org/10.3390/electronics10212679

Wei, W., & Mendel, J. M. (2000). Maximum-likelihood classification for digital amplitude-phase modulations. IEEE Transactions on Communications, 48(2), 189–193. https://doi.org/10.1109/26.823550

Panagiotou, P., Anastasopoulos, A., & Polydoros, A. (2000). Likelihood ratio tests for modulation classification. In MILCOM 2000 Proceedings. In 21st Century Military Communications. Architectures and Technologies for Information Superiority (Vol. 2, pp. 670–674). https://doi.org/10.1109/milcom.2000.904013

Wu, H.-C., Saquib, M., & Yun, Z. (2008). Novel automatic modulation classification using cumulant features for communications via multipath channels. IEEE Transactions on Wireless Communications, 7(8), 3098–3105. https://doi.org/10.1109/twc.2008.070015

Ramkumar, B. (2009). Automatic modulation classification for cognitive radios using cyclic feature detection. IEEE Circuits and Systems Magazine, 9(2), 27–45. https://doi.org/10.1109/mcas.2008.931739

Pawar, S. U., & Doherty, J. F. (2011). Modulation recognition in continuous phase modulation using approximate entropy. IEEE Transactions on Information Forensics and Security, 6(3), 843–852. https://doi.org/10.1109/tifs.2011.2159000

Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30, 5998–6008.

Srinivas, A., Lin, T.-Y., Parmar, N., Shlens, J., Abbeel, P., & Vaswani, A. (2021). Bottleneck transformers for visual recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 16519–16529). https://ieeexplore.ieee.org/abstract/document/9577771

Lin, S., Zeng, Y., & Gong, Y. (2022). Learning of time-frequency attention mechanism for automatic modulation recognition. IEEE Wireless Communications Letters, 11(4), 707–711. https://doi.org/10.1109/lwc.2022.3140828

Zhang, W., Sun, Y., Xue, K., & Yao, A. (2023). Research on modulation recognition algorithm based on channel and spatial self-attention mechanism. IEEE Access, 11, 68617–68631. https://doi.org/10.1109/access.2023.3292408

Liang, Z., Tao, M., Xie, J., Yang, X., & Wang, L. (2022). A radio signal recognition approach based on complex-valued CNN and self-attention mechanism. IEEE Transactions on Cognitive Communications and Networking, 8(3), 1358–1373. https://doi.org/10.1109/tccn.2022.3179450

Qi, M., Shi, N., Wang, G., & Shao, H. (2024). Data-transform multi-channel hybrid deep learning for automatic modulation recognition. IEEE Access, 12, 59113–59121. https://doi.org/10.1109/access.2024.3393481

Feng, Y., Duan, R., Li, S., Cheng, P., & Liu, W. (2025). A dual-branch network with feature assistance for automatic modulation recognition. IEEE Signal Processing Letters, 32, 701–705. https://doi.org/10.1109/lsp.2025.3527901

Zhang, W., Xue, K., Yao, A., & Sun, Y. (2024). CTRNet: An automatic modulation recognition based on transformer-CNN neural network. Electronics, 13(17), 3408. https://doi.org/10.3390/electronics13173408

Duan, R., Zhang, S., Wang, X., Li, Y., Liu, W., & Feng, Y. (2023). A multi-modal modulation recognition method with SNR segmentation based on time domain signals and constellation diagrams. Electronics, 12(14), 3175. https://doi.org/10.3390/electronics12143175

Luo, Z., Xiao, W., Zhang, X., Zhu, L., & Xiong, X. (2024). RLITNN: A multi-channel modulation recognition model combining multi-modal features. IEEE Transactions on Wireless Communications, 23, 1–15. https://doi.org/10.1109/twc.2024.3478752

Kong, W., Jiao, X., Xu, Y., Zhang, B., & Yang, Q. (2023). A transformer-based contrastive semi-supervised learning framework for automatic modulation recognition. IEEE Transactions on Cognitive Communications and Networking, 9(4), 950–962. https://doi.org/10.1109/tccn.2023.3264908

O’Shea, T. J., Roy, T., & Clancy, T. C. (2018). Over-the-air deep learning based radio signal classification. IEEE Journal of Selected Topics in Signal Processing, 12(1), 168–179. https://doi.org/10.1109/jstsp.2018.2797022

O’Shea, T. J., & West, N. (2016). Radio machine learning dataset generation with GNU Radio. Proceedings of the GNU Radio Conference, 1(1). https://pubs.gnuradio.org/index.php/grcon/article/view/11

Xu, J., Luo, C., Parr, G., & Luo, Y. (2020). A spatiotemporal multi-channel learning framework for automatic modulation recognition. IEEE Wireless Communications Letters, 9(10), 1629–1632. https://doi.org/10.1109/lwc.2020.2999453

Perenda, E., Rajendran, S., & Pollin, S. (2019). Automatic modulation classification using parallel fusion of convolutional neural networks. IEEE Transactions on Vehicular Technology, 69(9), 9825–9837. https://doi.org/10.1109/tvt.2020.3000148

Huynh-The, T., Hua, C.-H., Pham, Q.-V., & Kim, D.-S. (2020). MCNet: An efficient CNN architecture for robust automatic modulation classification. IEEE Communications Letters, 24(4), 811–815. https://doi.org/10.1109/lcomm.2020.2968030

Zhang, F., Luo, C., Xu, J., Luo, Y., & Zheng, F.-C. (2022). Deep learning based automatic modulation recognition: Models, datasets, and challenges. Digital Signal Processing, 129, 103650. https://doi.org/10.1016/j.dsp.2022.103650