Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Sep 27;14(1):22401.
doi: 10.1038/s41598-024-73547-w.

A complex-valued convolutional fusion-type multi-stream spatiotemporal network for automatic modulation classification

Yuying Wang  1 , Shengliang Fang  2 , Youchen Fan  3 , Mengtao Wang  1 , Zhaojing Xu  1 , Shunhu Hou  1
Affiliations

A complex-valued convolutional fusion-type multi-stream spatiotemporal network for automatic modulation classification

Yuying Wang et al. Sci Rep. .

Abstract

Automatic Modulation Classification (AMC) is crucial in non-cooperative communication systems as it facilitates the identification of interference signals with minimal prior knowledge. Although there have been significant advancements in Deep Learning (DL) within the field of AMC, leveraging the inherent relationships between In-phase (I) and Quadrature-phase (Q) components, and enhance recognition accuracy under low signal-to-noise ratio (SNR) conditions remains a challenge. This study introduces a complex-valued convolutional fusion-type multi-stream spatiotemporal network (CC-MSNet) for AMC, which combines spatial and temporal feature extraction modules for modulation recognition. Experimental results demonstrate that the CC-MSNet performs well on three benchmark datasets, RML2016.10a, RML2016.10b, and RML2016.04c, with average recognition accuracy of 62.86%, 65.08%, and 71.12%. It also performs excellently in low SNR environments below 0dB, significantly outperforming other networks.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
The modulation and demodulation process.
Fig. 2
Fig. 2
CC-MSNet network architecture.
Fig. 3
Fig. 3
Complex number convolution process.
Fig. 4
Fig. 4
Multi-stream spatial feature extraction module.
Fig. 5
Fig. 5
Recognition framework of LSTM.
Fig. 6
Fig. 6
Recognition framework of Bi-LSTM.
Fig. 7
Fig. 7
Time feature extraction module.
Fig. 8
Fig. 8
Comparison experiment of CC-MSNet spatial feature extraction module network.
Fig. 9
Fig. 9
Comparison experiment of CC-MSNet time feature extraction module network.
Fig. 10
Fig. 10
Classification accuracy of different networks on RML2016.10a.
Fig. 11
Fig. 11
Classification accuracy of different networks on RML2016.10b.
Fig. 12
Fig. 12
Classification accuracy of different networks on RML2016.04c.
Fig. 13
Fig. 13
Comparison of the training/validation curves for CC-MSNet and MCLDNN.
Fig. 14
Fig. 14
(a1–f1) is confusion matrices of the proposed and benchmark models on RML2016.10a; (a2–f2) is confusion matrices on RML2016.10b; (a3–f3) is confusion matrices on RML2016.04c.
Fig. 14
Fig. 14
(a1–f1) is confusion matrices of the proposed and benchmark models on RML2016.10a; (a2–f2) is confusion matrices on RML2016.10b; (a3–f3) is confusion matrices on RML2016.04c.
See this image and copyright information in PMC

References

    1. Guo, K., Dong, C. & An, K.NOMA-based cognitive satellite terrestrial relay network: secrecy performance under channel estimation errors and hardware impairments.IEEE Internet Things J.9(18), 17334–17347. 10.1109/JIOT.2022.3157673 (2022).
    1. Guo, K., Li, X., Alazab, M., Jhaveri, R. H. & An, K. Integrated satellite multiple two-way relay networks: secrecy performance under multiple eves and vehicles with non-ideal hardware. IEEE Trans. Intell. Veh.10.1109/TIV.2022.3215011 (2022).
    1. O’Shea, T. J., Corgan, J. & Clancy, T. C. Convolutional radio modulation recognition networks. In Engineering Applications of Neural Networks: 17th International Conference, EANN, Aberdeen, UK, September 2–5, 2016, Proceedings 17, 213–226 10.1007/978-3-319-44188-7_16 (Springer, 2016).
    1. West, N. E. & O’shea, T. Deep architectures for modulation recognition. In 2017 IEEE International Symposium on Dynamic Spectrum Access Networks (Dy SPAN), Piscataway, NJ, 2017, 1–6. 10.1109/DySPAN.2017.7920754 (2017).
    1. O’Shea, T. J., Roy, T. & Clancy, T. C. Over-the-air deep learning based radio signal classification. IEEE J. Sel. Top. Signal Process.12, 168–179. 10.1109/JSTSP.2018.2797022 (2018).

LinkOut - more resources