深度学习使能的自动调制分类技术研究进展

郑庆河; 李秉霖; 于治国; 姜蔚蔚; 朱政宇; 许驰; 黄崇文; 桂冠

doi:10.11999/JEIT250674

深度学习使能的自动调制分类技术研究进展

doi: 10.11999/JEIT250674 cstr: 32379.14.JEIT250674

1.
山东管理学院智能工程学院济南 250357
2.
北京邮电大学信息与通信工程学院北京 100876
3.
郑州大学电气与信息工程学院郑州 450001
4.
中国科学院沈阳自动化研究所沈阳 110034
5.
浙江大学信息与电子工程学院杭州 310058
6.
南京邮电大学通信与信息工程学院南京 210023

基金项目: 国家自然科学基金(62401070)，山东省重点研发计划(2024TSGC0055)，山东省自然科学基金(ZR2023QF125)，山东省高等学校青年创新团队计划(2024KJH005)

详细信息

作者简介:
郑庆河：男，副教授，研究方向为人工智能、信号处理、模式识别、无线通信

李秉霖：男，研究方向为深度学习、故障诊断、信号分析

于治国：男，教授，研究方向为人工智能、信号分析、工业互联网、物联网

姜蔚蔚：男，讲师，研究方向为物联网、人工智能、深度学习、信号处理

朱政宇：男，副教授，研究方向为无线通信、机器学习、智能反射面辅助通信、毫米波通信

许驰：男，研究员，研究方向为工业互联网、5G/6G、边缘计算

黄崇文：男，教授，研究方向为智能通感算、无线通信、人工智能

桂冠：男，教授，研究方向为智能通信、深度学习、认知计算

通讯作者:
郑庆河　zqh@sdmu.edu.cn

中图分类号: TN929.5
计量
- 文章访问数: 24
- HTML全文浏览量: 10
- PDF下载量: 8
- 被引次数: 0
出版历程
- 收稿日期: 2025-07-17
- 修回日期: 2025-09-30
- 网络出版日期: 2025-10-22

Research Progress of Deep Learning Enabled Automatic Modulation Classification Technology

1.
School of Intelligent Engineering, Shandong Management University, Jinan 250357, China
2.
School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing 100876, China
3.
School of Electrical and Information Engineering, Zhengzhou University, Zhengzhou 450001, China
4.
Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110034, China
5.
School of Information and Electronic Engineering, Zhejiang University, Hangzhou 310058, China
6.
School of Communication and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210023, China

Funds: The National Natural Science Foundation of China (62401070), The Shandong Provincial Key R&D Program (2024TSGC0055), The Shandong Provincial Natural Science Foundation (ZR2023QF125), The Shandong Provincial Youth Innovation Team Plan of Higher Education Institutions (2024KJH005)

摘要

摘要: 随着第六代无线通信系统向太赫兹频段以及空天地海一体化网络发展，通信环境呈现出高度异构化和超密集化的趋势，对自动调制识别技术提出了亚符号周期级别的精度要求。在复杂信道条件下，自动调制识别技术面临着时变多径信道引起的特征混叠、低信噪比环境下传统方法识别性能衰减以及稀疏码多址技术引发的混合调制信号检测复杂性提升等多重挑战。基于上述技术难题，该文从通信系统的信号传输特性出发，探讨自动调制分类方法设计的关键约束，系统回顾了深度学习使能的自动调制分类技术，综述了不同应用场景下自动调制分类方法面临的挑战，对经典深度学习模型进行了性能评估，最后概述了自动调制分类存在的问题及未来关键研究方向。
- 无线通信 /
- 深度学习 /
- 自动调制分类 /
- 时变多径信道
Abstract: Significance With the advancement of sixth-generation (6G) wireless communication systems towards the terahertz frequency band and space–air–ground integrated networks, the communication environment is becoming increasingly heterogeneous and densely deployed. This evolution imposes stringent precision requirements at the sub-symbol period level for Automatic Modulation Classification (AMC). Under complex channel conditions, AMC faces several challenges: feature mixing and distortion caused by time-varying multipath channels, substantial degradation in recognition accuracy of traditional methods under low Signal-to-Noise Ratio (SNR) conditions, and elevated complexity in detecting mixed modulation signals introduced by Sparse Code Multiple Access (SCMA) techniques. Addressing these challenges, this paper first analyzes the fundamental constraints on AMC method design from the perspective of signal transmission characteristics in communication models. It then systematically reviews Deep Learning (DL)-based AMC approaches, summarizes the difficulties these methods encounter in different wireless communication scenarios, evaluates the performance of representative DL models, and concludes with a discussion of current limitations in AMC together with promising research directions. Process Current research on AMC technology under complex channel conditions mainly focuses on three methodological categories: Likelihood-Based (LB), Feature-Based (FB), and DL, emphasizing both theoretical exploration and algorithmic innovation. Among these, end-to-end DL approaches have demonstrated superior performance in AMC tasks. By stacking multiple layers of nonlinear activation functions, DL models establish strong nonlinear fitting capabilities that allow them to uncover hidden patterns in radio signals. This enables DL to achieve high robustness and accuracy in complex environments. Convolutional Neural Networks (CNNs), leveraging their hierarchical local perception mechanism, can effectively capture amplitude and phase distortion characteristics of modulated signals, showing distinctive advantages in spatial feature extraction. Recurrent Neural Networks (RNNs), through the temporal memory function of gated units, exhibit theoretical superiority in modeling dynamic signal impairments such as inter-symbol interference, carrier frequency offset, carrier phase offset, and timing errors. More recently, Transformer architectures have achieved global feature association modeling through self-attention mechanisms, thereby enhancing the ability to identify key features and markedly improving AMC accuracy under low SNR conditions. The application potential of Transformers in AMC can be further extended by integrating multi-scale feature fusion, optimizing computational efficiency, and improving generalization. Prospects With the continuous growth of communication demands and the increasing complexity of application scenarios, the efficient and reliable management and utilization of wireless spectrum resources has become a central research focus. AMC enables mobile communication systems to achieve dynamic channel adaptation and heterogeneous network integration. Driven by the development of space–air–ground integrated networks, the application scope of AMC has expanded beyond traditional terrestrial cellular systems to emerging domains such as satellite communication and vehicular networking. DL-based AMC frameworks can capture dynamic channel responses through joint time–frequency domain representations, enhance transient feature extraction via attention mechanisms, and effectively decouple the coupling effects of multipath fading and Doppler shifts. By applying neural architecture search and model quantization–compression techniques, DL models can achieve low-complexity, real-time inference at the edge, thereby supporting end-to-end latency control in Vehicle-to-Everything (V2X) communication links. Furthermore, advanced DL architectures introduce feature enhancement mechanisms to preserve signal phase integrity, improving resilience against channel distortion. In dynamic optical network monitoring, feature extraction networks tailored to time-varying channels can adaptively capture the evolution of nonlinear phase shifts. Through implicit channel compensation, DL enables collaborative learning of time-domain and frequency-domain features. At present, AMC technology is progressing towards elastic architectures that support dynamic reconstruction of model parameters through online knowledge distillation and meta-learning frameworks, offering adaptive and lightweight solutions for Internet-of-Things (IoT) scenarios. Conclusions This paper systematically reviews the current research and challenges of AMC technology in the context of 6G networks. First, the applications of CNNs, RNNs, Transformers, and hybrid DL models in AMC are discussed in detail, with analysis of the technical advantages and limitations of each approach. Next, three representative application scenarios are examined: the mobile communication, the optical communication, and the IoT, highlighting the specific challenges faced by AMC technology. At present, the development of DL-driven AMC has moved beyond model design to include deployment and application challenges in real wireless communication environments. For example, constructing DL architectures with continuous learning capabilities is essential for adapting to dynamic communication conditions, while developing large-scale DL models provides an effective way to improve cross-scenario generalization. Future research should emphasize directions that integrate prior knowledge of the physical layer with DL architectures, strengthen feature fusion strategies, and advance hardware–algorithm co-design frameworks.
- Wireless communication /
- Deep Learning (DL) /
- Automatic Modulation Classification (AMC) /
- Time-varying multipath channel

HTML全文

图 1 无线通信系统模型

下载: 全尺寸图片幻灯片

图 2 典型CNN模型结构

下载: 全尺寸图片幻灯片

图 3 典型LSTM单元

下载: 全尺寸图片幻灯片

图 4 典型GRU单元

下载: 全尺寸图片幻灯片

图 5 用于AMC的Transformer模型结构

下载: 全尺寸图片幻灯片

图 6 AMC系统应用场景

下载: 全尺寸图片幻灯片

表 1 典型AMC方法性能对比

AMC方法	候选调制方式	信噪比(dB)	准确率(%)
P-LSTM^[24]	{B/Q/8PSK, 16/32QAM}	0:5:20	99.30 at 5 dB
3D CNN^[26]	{B/Q/8PSK, 16QAM}	–15:5:30	> 95.00 at 0 dB
可分离CNN^[29]	{2/4/8FSK, B/Q/8PSK, 16QAM}	–10:2:20	83.44
分布式CNN^[31]	RadioML 2016.10a	–20:2:18	平均62.41
信噪比分类网络^[32]	RadioML 2016.10a	–20:2:18	86.00 at 0 dB
CNN^[33]	RadioML 2018.01a	0:2:30	> 90.00 at 0 dB
CNN^[34]	RadioML 2016.10a	–20:2:18	平均63.8
CNN^[35]	RadioML 2016.10a	–20:2:18	> 85.00 at 0 dB
MWD-CNN^[36]	RadioML 2016.10a	–20:2:18	平均62.52
CNN^[37]	{BPSK, QPSK, 16QAM, 64QAM, GFSK, CPFSK}	30	97.00
ANR CNN^[38]	RadioML 2016.10a	–20:2:18	平均61.75
MobileNetV2^[39]	RadioML 2016.10a	20	95.00
ResNet-LSTM^[41]	RadioML 2016.10b	–20:2:18	> 90.00 at 0 dB
CBAM-GRU^[43]	RadioML 2016.10a	–20:2:18	> 92.79 at 0 dB
GR-ResNet^[44]	RadioML 2016.10a	–6:2:18	平均83.57
多注意力ResNet-GRU^[45]	RadioML 2016.10b	–20:2:18	平均64.78
LSTM-Transformer^[46]	RadioML 2016.10a	–20:2:18	> 89.90 at 0 dB
ResNet-Transformer^[48]	RadioML 2018.01a	–20:2:30	平均62.61
Bottleneck Transformer^[49]	{PAM2, PAM4, PAM8, PAM16, BPSK, QPSK}	8:2:30	99.16 at 8 dB
Transformer^[50]	RadioML 2016.10a	0:2:18	平均91.90
CNN-Transformer^[51]	RadioML 2016.10b	–20:2:18	平均64.01
Meta-Transformer^[52]	RadioML 2018.01a	–20:2:30	95.76 at 20 dB
稀疏Transformer^[71]	RadioML 2016.10a, RML22	–20:2:20	平均71.13
KAN-CNN^[70]	RadioML 2018.01a	–20:2:30	平均65.56
Ultra Lite CNN^[69]	RadioML 2016.10a	–20:2:18	平均62.47
PCTNet^[72]	RadioML 2016.10b	–20:2:18	平均64.60
双流Transformer^[73]	{OFDM, 2/4/8FSK, B/Q/8PSK, 16/64QAM}	–10:2:20	> 85.00 at 0 dB
自适应小波网络^[74]	RadioML 2016.10a/b, RadioML 2018.01a	–20:2:18/-20:2:30	平均62.44
时空CNN^[75]	RadioML 2016.10a	–20:2:18	平均61.50
多端CNN^[76]	{B/Q/OQ/8PSK, 16/64QAM, 16/32APSK}	0:1:10	95.00 at 5 dB
HKDD^[53]	HKDD_AMC12, HKDD_AMC36	–20:2:20/-20:2:30	> 90.00 at 20 dB
GGCNN^[54]	RadioML 2018.01a	–20:2:30	> 80.00 at 20 dB
ResSwinT–SwinT^[58]	{CW, SIN/LFM, B/4FSK, B/QPSK, polyphase codes}	–14:2:10	99.80 at 0 dB
DL-EVT^[61]	RadioML 2018.01a	0:2:30	> 90.00 at 8 dB
图卷积神经网络^[68]	{2/4ASK, 2/4FSK, B/QPSK, 16/64QAM}	–14:2:10	> 70.00 at 2 dB
对比自监督学习^[77]	RadioML 2018.01a	–20:2:30	49.50 at -5 dB
阈值去噪RNN^[78]	RadioML 2018.01a	–20:2:30	平均63.50
GIGNet^[79]	RadioML 2016.10a	–20:2:18	平均63.83
多尺度自适应小波分解网络^[80]	HisarMod 2019.1	–20:2:18	平均74.40
DCTFANet^[81]	RadioML 2016.10a	–20:2:18	平均71.60
卷积双注意力Transformer^[82]	RadioML 2018.01a	–20:2:30	平均63.86
M²–Net^[83]	Multiband signals	–20:2:10	平均51.17
自监督对齐数据增强网络^[84]	RadioML 2016.10a	–10:2:0	平均72.19
KG-MTSNet^[85]	RadioML 2016.10a	–6:2:18	平均86.63
FTPNet^[86]	RadioML 2018.01a	–20:2:30	平均56.44

下载: 导出CSV

参考文献(86)

[1]	SHAFIE A, YANG Nan, HAN Chong, et al. Terahertz communications for 6G and beyond wireless networks: Challenges, key advancements, and opportunities[J]. IEEE Network, 2023, 37(3): 162–169. doi: 10.1109/MNET.118.2200057.
[2]	YOSHIDA M, KASAI K, HIROOKA T, et al. Dual-polarization on-line 256 and 512 QAM digital coherent transmission[C]. Optical Fiber Communications Conference and Exhibition (OFC), San Diego, USA, 2019: 1–3. doi: 10.1364/OFC.2019.M2H.4.
[3]	HANNA S, DICK C, and CABRIC D. Signal processing-based deep learning for blind symbol decoding and modulation classification[J]. IEEE Journal on Selected Areas in Communications, 2022, 40(1): 82–96. doi: 10.1109/JSAC.2021.3126088.
[4]	LIU Mingqian, QU Nan, SHANG Bodong, et al. Energy and spectrum efficient blind equalization with unknown constellation for air-to-ground multipath UAV communications[J]. IEEE Transactions on Green Communications and Networking, 2021, 5(3): 1357–1368. doi: 10.1109/TGCN.2021.3073914.
[5]	WANG Yang, ZHOU Xue, LIAO Xi, et al. Channel measurements and multipath characterization for indoor sub-terahertz communication[J]. IEEE Transactions on Vehicular Technology, 2025, 74(3): 4393–4407. doi: 10.1109/TVT.2024.3493119.
[6]	JIANG Hao, SHI Wangqi, CHEN Zhen, et al. Dynamic channel modeling of fluid antenna systems in UAV communications[J]. IEEE Wireless Communications Letters, 2025, 14(10): 3169–3173. doi: 10.1109/LWC.2025.3588223.
[7]	JIANG Hao, SHI Wangqi, ZHANG Zaichen, et al. Large-scale RIS enabled air-ground channels: Near-field modeling and analysis[J]. IEEE Transactions on Wireless Communications, 2025, 24(2): 1074–1088. doi: 10.1109/TWC.2024.3504839.
[8]	REBHI M, HASSAN K, RAOOF K, et al. Sparse code multiple access: Potentials and challenges[J]. IEEE Open Journal of the Communications Society, 2021, 2: 1205–1238. doi: 10.1109/OJCOMS.2021.3081166.
[9]	CHEN Zhen, GUO Yeyong, ZHANG Peichang, et al. Physical layer security improvement for hybrid RIS-assisted MIMO communications[J]. IEEE Communications Letters, 2024, 28(11): 2493–2497. doi: 10.1109/LCOMM.2024.3427010.
[10]	PANAGIOTOU P, ANASTASOPOULOS A, and POLYDOROS A. Likelihood ratio tests for modulation classification[C]. The 21st Century Military Communications. Architectures and Technologies for Information Superiority, Los Angeles, USA, 2000: 670–674. doi: 10.1109/MILCOM.2000.904013.
[11]	闫文康, 闫毅, 范亚楠, 等. 基于小波变换熵值及高阶累积量联合的卫星信号调制识别算法[J]. 空间科学学报, 2021, 41(6): 968–975. doi: 10.11728/cjss2021.06.968. YAN Wenkang, YAN Yi, FAN Yanan, et al. A modulation recognition algorithm based on wavelet transform entropy and high-order cumulant for satellite signal modulation[J]. Chinese Journal of Space Science, 2021, 41(6): 968–975. doi: 10.11728/cjss2021.06.968.
[12]	陈啸锋, 张茜茜, 桂冠. 基于渐进式神经架构搜索的自动调制分类方法[J]. 太赫兹科学与电子信息学报, 2024, 22(11): 1289–1295. doi: 10.11805/TKYDA2023080. CHEN Xiaofeng, ZHANG Xixi, and GUI Guan. Progressive neural architecture search based automatic modulation classification method[J]. Journal of Terahertz Science and Electronic Information Technology, 2024, 22(11): 1289–1295. doi: 10.11805/TKYDA2023080.
[13]	王栋, 崔天舒, 姬丽彬, 等. 基于迁移学习的自动调制分类方法[J/OL]. 北京航空航天大学学报, https://doi.org/10.13700/j.bh.1001-5965.2024.0231, 2024. WANG Dong, CUI Tianshu, JI Libin, et al. Automatic modulation classification based on transfer learning[J/OL]. Journal of Beijing University of Aeronautics and Astronautics, https://doi.org/10.13700/j.bh.1001-5965.2024.0231, 2024.
[14]	ZHENG Qinghe, TIAN Xinyu, YU Zhiguo, et al. Robust automatic modulation classification using asymmetric trilinear attention net with noisy activation function[J]. Engineering Applications of Artificial Intelligence, 2025, 141: 109861. doi: 10.1016/j.engappai.2024.109861.
[15]	ZHENG Qinghe, TIAN Xinyu, YU Zhiguo, et al. DL-PR: Generalized automatic modulation classification method based on deep learning with priori regularization[J]. Engineering Applications of Artificial Intelligence, 2023, 122: 106082. doi: 10.1016/j.engappai.2023.106082.
[16]	TEKBıYıK K, EKTI A R, GÖRÇIN A, et al. Robust and fast automatic modulation classification with CNN under multipath fading channels[C]. IEEE 91st Vehicular Technology Conference (VTC2020-Spring), Antwerp, Belgium, 2020: 1–6. doi: 10.1109/VTC2020-Spring48590.2020.9128408.
[17]	LUO Zhongqiang, XIAO Wenshi, ZHANG Xueqin, et al. RLITNN: A multi-channel modulation recognition model combining multi-modal features[J]. IEEE Transactions on Wireless Communications, 2024, 23(12): 19083–19097. doi: 10.1109/TWC.2024.3478752.
[18]	MENG Fan, CHEN Peng, WU Lenan, et al. Automatic modulation classification: A deep learning enabled approach[J]. IEEE Transactions on Vehicular Technology, 2018, 67(11): 10760–10772. doi: 10.1109/TVT.2018.2868698.
[19]	CHEN Yufan, SHAO Wei, LIU Jin, et al. Automatic modulation classification scheme based on LSTM with random erasing and attention mechanism[J]. IEEE Access, 2020, 8: 154290–154300. doi: 10.1109/ACCESS.2020.3017641.
[20]	ZHENG Qinghe, ZHAO Penghui, LI Yang, et al. Spectrum interference-based two-level data augmentation method in deep learning for automatic modulation classification[J]. Neural Computing and Applications, 2021, 33(13): 7723–7745. doi: 10.1007/s00521-020-05514-1.
[21]	ZHANG Zufan, LUO Hao, WANG Chun, et al. Automatic modulation classification using CNN-LSTM based dual-stream structure[J]. IEEE Transactions on Vehicular Technology, 2020, 69(11): 13521–13531. doi: 10.1109/TVT.2020.3030018.
[22]	CAI Jinging, GAN Fengming, CAO Xianghai, 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.
[23]	HAMIDI-RAD S and JAIN S. MCformer: A transformer based deep neural network for automatic modulation classification[C]. IEEE Global Communications Conference (GLOBECOM), Madrid, Spain, 2021: 1–6. doi: 10.1109/GLOBECOM46510.2021.9685815.
[24]	HAO Xiaofeng, ZHANG Huadi, and GUO Rui. Digital modulation recognition based on high-order cumulants and P-LSTM[J]. KSII Transactions on Internet and Information Systems, 2024, 18(11): 3324–3338. doi: 10.3837/tiis.2024.11.013.
[25]	USSIPOV N, AKHTANOV S, ZHANABAEV Z, et al. Automatic modulation classification for MIMO system based on the mutual information feature extraction[J]. IEEE Access, 2024, 12: 68463–68470. doi: 10.1109/ACCESS.2024.3400448.
[26]	HUYNH-THE T, NGUYEN T V, PHAM Q V, et al. MIMO-OFDM modulation classification using three-dimensional convolutional network[J]. IEEE Transactions on Vehicular Technology, 2022, 71(6): 6738–6743. doi: 10.1109/TVT.2022.3159254.
[27]	BAI Jing, LIU Xuebo, WANG Yiran, et al. Integrating prior knowledge and contrast feature for signal modulation classification[J]. IEEE Internet of Things Journal, 2024, 11(12): 21461–21473. doi: 10.1109/JIOT.2024.3377916.
[28]	HUANG Sai, HE Jiashuo, YANG Zheng, et al. Generalized automatic modulation classification for OFDM systems under unseen synthetic channels[J]. IEEE Transactions on Wireless Communications, 2024, 23(9): 11931–11941. doi: 10.1109/TWC.2024.3386762.
[29]	FU Xue, GUI Guan, WANG Yu, et al. Lightweight automatic modulation classification based on decentralized learning[J]. IEEE Transactions on Cognitive Communications and Networking, 2022, 8(1): 57–70. doi: 10.1109/TCCN.2021.3089178.
[30]	ZHANG Haozheng, HUANG Ming, YANG Jingjing, et al. A data preprocessing method for automatic modulation classification based on CNN[J]. IEEE Communications Letters, 2021, 25(4): 1206–1210. doi: 10.1109/LCOMM.2020.3044755.
[31]	杨洁, 董标, 付雪, 等. 基于轻量化分布式学习的自动调制分类方法[J]. 通信学报, 2022, 43(7): 134–142. doi: 10.11959/j.issn.1000-436x.2022145. YANG Jie, DONG Biao, FU Xue, et al. Lightweight decentralized learning-based automatic modulation classification method[J]. Journal on Communications, 2022, 43(7): 134–142. doi: 10.11959/j.issn.1000-436x.2022145.
[32]	郭业才, 姚文强. 基于信噪比分类网络的调制信号分类识别算法[J]. 电子与信息学报, 2022, 44(10): 3507–3515. doi: 10.11999/JEIT210825. GUO Yecai and YAO Wenqiang. Modulation signal classification and recognition algorithm based on signal to noise ratio classification network[J]. Journal of Electronics & Information Technology, 2022, 44(10): 3507–3515. doi: 10.11999/JEIT210825.
[33]	OIKONOMOU T K, EVGENIDIS N G, NIXARLIDIS D G, et al. CNN-based automatic modulation classification under phase imperfections[J]. IEEE Wireless Communications Letters, 2024, 13(5): 1508–1512. doi: 10.1109/LWC.2024.3379198.
[34]	MOHSEN S, ALI A M, and EMAM A. Automatic modulation recognition using CNN deep learning models[J]. Multimedia Tools and Applications, 2024, 83(3): 7035–7056. doi: 10.1007/s11042-023-15814-y.
[35]	CHAHIL S T H, ZAKWAN M, KHAN K, et al. Performance analysis of different signal representations and optimizers for CNN based automatic modulation classification[J]. Wireless Personal Communications, 2024, 139(4): 2503–2528. doi: 10.1007/s11277-024-11722-y.
[36]	龚安, 张贵临, 牟伟清, 等. 基于多层小波分解卷积神经网络的自动调制识别方法[J/OL]. 无线电通信技术, https://link.cnki.net/urlid/13.1099.TN.20241125.1432.004, 2024. GONG An, ZHANG Guilin, MU Weiqing, et al. Automatic modulation recognition method based on multi-layer wavelet decomposition convolutional neural network[J/OL]. Radio Communications Technology, https://link.cnki.net/urlid/13.1099.TN.20241125.1432.004, 2024.
[37]	袁博文, 秦怀涛, 易卫明, 等. 基于卷积神经网络的数字调制分类识别[J]. 无线通信技术, 2023, 32(3): 54–57,62. doi: 10.3969/j.issn.1003-8329.2023.03.011. YUAN Bowen, QIN Huaitao, YI Weiming, et al. Classification and recognition of digital modulation based on convolutional neural network[J]. Wireless Communication Technology, 2023, 32(3): 54–57,62. doi: 10.3969/j.issn.1003-8329.2023.03.011.
[38]	陈昊, 郭文普, 康凯, 等. 基于卷积自适应降噪网络的自动调制识别方法[J]. 无线电工程, 2025, 55(2): 291–297. doi: 10.3969/j.issn.1003-3106.2025.02.008. CHEN Hao, GUO Wenpu, KANG Kai, et al. Automatic modulation recognition method based on convolutional adaptive denoising network[J]. Radio Engineering, 2025, 55(2): 291–297. doi: 10.3969/j.issn.1003-3106.2025.02.008.
[39]	ZHENG Qinghe, SAPONARA S, TIAN Xinyu, et al. A real-time constellation image classification method of wireless communication signals based on the lightweight network MobileViT[J]. Cognitive Neurodynamics, 2024, 18(2): 659–671. doi: 10.1007/s11571-023-10015-7.
[40]	HAMZA M A, HASSINE S B H, LARABI-MARIE-SAINTE S, et al. Optimal bidirectional LSTM for modulation signal classification in communication systems[J]. Computers, Materials & Continua, 2022, 72(2): 3055–3071. doi: 10.32604/cmc.2022.024490.
[41]	ELSAGHEER M M and RAMZY S M. A hybrid model for automatic modulation classification based on residual neural networks and long short term memory[J]. Alexandria Engineering Journal, 2023, 67: 117–128. doi: 10.1016/j.aej.2022.08.019.
[42]	RAJENDRAN S, MEERT W, GIUSTINIANO D, et al. Deep learning models for wireless signal classification with distributed low-cost spectrum sensors[J]. IEEE Transactions on Cognitive Communications and Networking, 2018, 4(3): 433–445. doi: 10.1109/TCCN.2018.2835460.
[43]	杨宵, 姚爱琴, 孙运强, 等. 基于CBAM-GRU的通信信号自动调制识别[J]. 遥测遥控, 2024, 45(5): 73–81. doi: 10.12347/j.ycyk.20240606002. YANG Xiao, YAO Aiqin, SUN Yunqiang, et al. Automatic modulation and recognition of communication signals based on CBAM-GRU[J]. Journal of Telemetry, Tracking and Command, 2024, 45(5): 73–81. doi: 10.12347/j.ycyk.20240606002.
[44]	HUANG Sai, DAI Rui, HUANG Juanjuan, et al. Automatic modulation classification using gated recurrent residual network[J]. IEEE Internet of Things Journal, 2020, 7(8): 7795–7807. doi: 10.1109/JIOT.2020.2991052.
[45]	李鸣皓, 解志斌, 颜培玉, 等. 基于多注意力残差网络和GRU的自动调制识别算法[J]. 无线电工程, 2025, 55(1): 36–44. doi: 10.3969/j.issn.1003-3106.2025.01.005. LI Minghao, XIE Zhibin, YAN Peiyu, et al. Automatic modulation recognition algorithm based on multi-attention residual network and GRU[J]. Radio Engineering, 2025, 55(1): 36–44. doi: 10.3969/j.issn.1003-3106.2025.01.005.
[46]	ZHANG Ziwei, ZHU Mengtao, LI Yunjie, et al. Joint recognition and parameter estimation of cognitive radar work modes with LSTM-transformer[J]. Digital Signal Processing, 2023, 140: 104081. doi: 10.1016/j.dsp.2023.104081.
[47]	HOU Dongbin, LI Lixin, LIN Wensheng, et al. ClST: A convolutional transformer framework for automatic modulation recognition by knowledge distillation[J]. IEEE Transactions on Wireless Communications, 2024, 23(7): 8013–8028. doi: 10.1109/TWC.2023.3347537.
[48]	YING Shanchuan, HUANG Sai, CHANG Shuo, et al. A convolutional and transformer based deep neural network for automatic modulation classification[J]. China Communications, 2023, 20(5): 135–147. doi: 10.23919/JCC.ja.2022-0580.
[49]	梁坤, 刘战胜. 基于联合残差网络和Bottleneck Transformer的调制格式识别方法[J]. 光通信技术, 2024, 48(3): 13–17. doi: 10.13921/j.cnki.issn1002-5561.2024.03.003. LIANG Kun and LIU Zhansheng. Modulation format identification method based on joint residual network and Bottleneck Transformers[J]. Optical Communication Technology, 2024, 48(3): 13–17. doi: 10.13921/j.cnki.issn1002-5561.2024.03.003.
[50]	战权海, 张雄伟, 宋磊, 等. 基于改进Transformer的自动调制识别方法[J]. 数据采集与处理, 2024, 39(6): 1410–1419. doi: 10.16337/j.1004-9037.2024.06.010. ZHAN Quanhai, ZHANG Xiongwei, SONG Lei, et al. Automatic modulation recognition method based on improved Transformer[J]. Journal of Data Acquisition and Processing, 2024, 39(6): 1410–1419. doi: 10.16337/j.1004-9037.2024.06.010.
[51]	杨静雅, 齐彦丽, 周一青, 等. CNN-Transformer轻量级智能调制识别算法[J]. 西安电子科技大学学报, 2023, 50(3): 40–49. doi: 10.19665/j.issn1001-2400.2023.03.004. YANG Jingya, QI Yanli, ZHOU Yiqing, et al. Algorithm for recognition of lightweight intelligent modulation based on the CNN-transformer networks[J]. Journal of Xidian University, 2023, 50(3): 40–49. doi: 10.19665/j.issn1001-2400.2023.03.004.
[52]	JANG J, PYO J, YOON Y I, et al. Meta-transformer: A meta-learning framework for scalable automatic modulation classification[J]. IEEE Access, 2024, 12: 9267–9276. doi: 10.1109/ACCESS.2024.3352634.
[53]	ZHENG Shilian, ZHOU Xiaoyu, ZHANG Luxin, et al. Toward next-generation signal intelligence: A hybrid knowledge and data-driven deep learning framework for radio signal classification[J]. IEEE Transactions on Cognitive Communications and Networking, 2023, 9(3): 564–579. doi: 10.1109/TCCN.2023.3243899.
[54]	GHASEMZADEH P, HEMPEL M, WANG Honggang, et al. GGCNN: An efficiency-maximizing gated graph convolutional neural network architecture for automatic modulation identification[J]. IEEE Transactions on Wireless Communications, 2023, 22(9): 6033–6047. doi: 10.1109/TWC.2023.3239311.
[55]	CLANCY J, MULLINS D, DEEGAN B, et al. Wireless access for V2X communications: Research, challenges and opportunities[J]. IEEE Communications Surveys & Tutorials, 2024, 26(3): 2082–2119. doi: 10.1109/COMST.2024.3384132.
[56]	NOOR-A-RAHIM M, LIU Zilong, LEE H, et al. 6G for vehicle-to-everything (V2X) communications: Enabling technologies, challenges, and opportunities[J]. Proceedings of the IEEE, 2022, 110(6): 712–734. doi: 10.1109/JPROC.2022.3173031.
[57]	KIM S H, MOON C B, KIM J W, et al. A hybrid deep learning model for automatic modulation classification[J]. IEEE Wireless Communications Letters, 2022, 11(2): 313–317. doi: 10.1109/LWC.2021.3126821.
[58]	REN Bing, TEH K C, AN Hongyang, et al. Automatic modulation recognition of dual-component radar signals using ResSwinT–SwinT network[J]. IEEE Transactions on Aerospace and Electronic Systems, 2023, 59(5): 6405–6418. doi: 10.1109/TAES.2023.3277430.
[59]	王业恒, 吴彰, 赵永胜, 等. 一种用于可见光通信信号调制格式识别的改进YOLOv5s算法[J]. 光通信技术, 2024, 48(3): 18–22. doi: 10.13921/j.cnki.issn1002-5561.2024.03.004. WANG Yeheng, WU Zhang, ZHAO Yongsheng, et al. Improved YOLOv5s algorithm for modulation format recognition of visible light communication signal[J]. Optical Communication Technology, 2024, 48(3): 18–22. doi: 10.13921/j.cnki.issn1002-5561.2024.03.004.
[60]	CHENG En, YAN Jiaquan, SUN Haixin, et al. Research on MPSK modulation classification of underwater acoustic communication signals[C]. IEEE/OES China Ocean Acoustics (COA), Harbin, China, 2016: 1–5. doi: 10.1109/COA.2016.7535788.
[61]	CHEN Yanghong, XU Xiaodong, and QIN Xiaowei. An open-set modulation recognition scheme with deep representation learning[J]. IEEE Communications Letters, 2023, 27(3): 851–855. doi: 10.1109/LCOMM.2023.3241388.
[62]	刘高辉, 王壮壮. 基于轻量型网络的单载波信号调制识别[J]. 计算机系统应用, 2023, 32(8): 238–243. doi: 10.15888/j.cnki.csa.009213. LIU Gaohui and WANG Zhuangzhuang. Modulation recognition of single carrier signal based on lightweight network[J]. Computer Systems & Applications, 2023, 32(8): 238–243. doi: 10.15888/j.cnki.csa.009213.
[63]	马文轩, 蔡卓燃, 王川, 等. 基于轻量级混合神经网络的边缘设备调制识别方法[J]. 信息对抗技术, 2024, 3(6): 83–94. doi: 10.12399/j.issn.2097-163x.2024.06.008. MA Wenxuan, CAI Zhuoran, WANG Chuan, et al. Edge devices modulation recognition method based on lightweight hybrid neural network[J]. Information Countermeasure Technology, 2024, 3(6): 83–94. doi: 10.12399/j.issn.2097-163x.2024.06.008.
[64]	O’SHEA T J, CORGAN J, and CLANCY T C. Convolutional radio modulation recognition networks[C]. 17th International Conference on Engineering Applications of Neural Networks, Aberdeen, UK, 2016: 213–226. doi: 10.1007/978-3-319-44188-7_16.
[65]	O’SHEA T J, ROY T, and 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.
[66]	SATHYANARAYANAN V, GERSTOFT P, and EL GAMAL A. RML22: Realistic dataset generation for wireless modulation classification[J]. IEEE Transactions on Wireless Communications, 2023, 22(11): 7663–7675. doi: 10.1109/TWC.2023.3254490.
[67]	SNOAP J A, POPESCU D C, and SPOONER C M. Deep-learning-based classifier with custom feature-extraction layers for digitally modulated signals[J]. IEEE Transactions on Broadcasting, 2024, 70(3): 763–773. doi: 10.1109/TBC.2024.3391056.
[68]	LIU Yabo, LIU Yi, and YANG Cheng. Modulation recognition with graph convolutional network[J]. IEEE Wireless Communications Letters, 2020, 9(5): 624–627. doi: 10.1109/LWC.2019.2963828.
[69]	GUO Lantu, WANG Yu, LIU Yuchao, et al. Ultralight convolutional neural network for automatic modulation classification in internet of unmanned aerial vehicles[J]. IEEE Internet of Things Journal, 2024, 11(11): 20831–20839. doi: 10.1109/JIOT.2024.3373497.
[70]	郑庆河, 刘方霖, 余礼苏, 等. 基于改进Kolmogorov-Arnold混合卷积神经网络的调制识别方法[J]. 电子与信息学报, 2025, 47(8): 2584–2597. doi: 10.11999/JEIT250161. ZHENG Qinghe, LIU Fanglin, YU Lisu, et al. An improved modulation recognition method based on hybrid kolmogorov-arnold convolutional neural network[J]. Journal of Electronics & Information Technology, 2025, 47(8): 2584–2597. doi: 10.11999/JEIT250161.
[71]	郑庆河, 刘方霖, 余礼苏, 等. 一种结合小波去噪卷积与稀疏Transformer的调制识别方法[J]. 电子与信息学报, 2025, 47(7): 2361–2374. doi: 10.11999/JEIT241159. ZHENG Qinghe, LIU Fanglin, YU Lisu, et al. A modulation recognition method combining wavelet denoising convolution and sparse Transformer[J]. Journal of Electronics & Information Technology, 2025, 47(7): 2361–2374. doi: 10.11999/JEIT241159.
[72]	MA Wenxuan, CAI Zhuoran, and WANG Chuan. A transformer and convolution-based learning framework for automatic modulation classification[J]. IEEE Communications Letters, 2024, 28(6): 1392–1396. doi: 10.1109/LCOMM.2024.3380623.
[73]	LI Juan, JIA Qingning, CUI Xuerong, et al. Automatic modulation recognition of underwater acoustic signals using a two-stream transformer[J]. IEEE Internet of Things Journal, 2024, 11(10): 18839–18851. doi: 10.1109/JIOT.2024.3367852.
[74]	ZHANG Jiawei, WANG Tiantian, FENG Zhixi, et al. Toward the automatic modulation classification with adaptive wavelet network[J]. IEEE Transactions on Cognitive Communications and Networking, 2023, 9(3): 549–563. doi: 10.1109/TCCN.2023.3252580.
[75]	陈发堂, 刘泽, 范子健. 基于时空卷积网络的通信信号调制识别[J]. 电讯技术, 2025, 65(4): 518–524. doi: 10.20079/j.issn.1001-893x.240116003. CHEN Fatang, LIU Ze, and FAN Zijian. Modulation recognition of communication signals based on spatiotemporal convolutional network[J]. Telecommunication Engineering, 2025, 65(4): 518–524. doi: 10.20079/j.issn.1001-893x.240116003.
[76]	查雄, 彭华, 秦鑫, 等. 基于多端卷积神经网络的调制识别方法[J]. 通信学报, 2019, 40(11): 30–37. doi: 10.11959/j.issn.1000-436x.2019206. ZHA Xiong, PENG Hua, QIN Xin, et al. Modulation recognition method based on multi-inputs convolution neural network[J]. Journal on Communications, 2019, 40(11): 30–37. doi: 10.11959/j.issn.1000-436x.2019206.
[77]	DU Mingyang, PAN Jifei, and BI Daping. A contrastive learner for automatic modulation classification[J]. IEEE Transactions on Wireless Communications, 2025, 24(4): 3575–3589. doi: 10.1109/TWC.2025.3532438.
[78]	AN T T, ARGYRIOU A, PUSPITASARI A A, et al. Efficient automatic modulation classification for next-generation wireless networks[J]. IEEE Transactions on Green Communications and Networking, 2025. doi: 10.1109/TGCN.2025.3574278. (查阅网上资料,未找到本条文献卷期页码,请确认).
[79]	KE Yang, ZHANG Wancheng, ZHANG Yan, et al. GIGNet: A graph-in-graph neural network for automatic modulation recognition[J]. IEEE Transactions on Vehicular Technology, 2025, 74(6): 10058–10062. doi: 10.1109/TVT.2025.3542494.
[80]	QIN Xiaoqian, JIANG Weiheng, GUI Guan, et al. Multilevel adaptive wavelet decomposition network-based automatic modulation recognition: Exploiting time-frequency multiscale correlations[J]. IEEE Transactions on Cognitive Communications and Networking, 2025, 11(5): 3218–3231. doi: 10.1109/TCCN.2025.3535738.
[81]	FENG Yuhang, DUAN Ruifeng, LI Shurui, et al. A dual-branch network with feature assistance for automatic modulation recognition[J]. IEEE Signal Processing Letters, 2025, 32: 701–705. doi: 10.1109/LSP.2025.3527901.
[82]	YI Zengrui, MENG Hua, GAO Lu, et al. Efficient convolutional dual-attention transformer for automatic modulation recognition[J]. Applied Intelligence, 2025, 55(3): 231. doi: 10.1007/s10489-024-06202-6.
[83]	ZHANG Xingjian, WANG Pengxu, MA Yuan, et al. M²-Net: Multitask-learning-based multiband signal recognition network[J]. IEEE Internet of Things Journal, 2025, 12(11): 16543–16558. doi: 10.1109/JIOT.2025.3535744.
[84]	ZHANG Ziwei, LI Yunjie, ZHU Mengtao, et al. Self-supervised aligned data augmentation network for imbalanced modulation classification[J]. IEEE Internet of Things Journal, 2025, 12(15): 30862–30878. doi: 10.1109/JIOT.2025.3571448.
[85]	LI Yike, ZHOU Fuhui, YUAN Lu, et al. A novel knowledge graph driven automatic modulation classification framework for 6G wireless communications[J]. IEEE Transactions on Wireless Communications, 2025, 24(3): 2373–2388. doi: 10.1109/TWC.2024.3520661.
[86]	FENG Shuai, WANG Yatong, WEN Zhongyi, et al. Fine-grained transductive prototypical network based few-shot signal modulation classification using coarse labels[J]. IEEE Transactions on Cognitive Communications and Networking, 2025. doi: 10.1109/TCCN.2025.3594331. (查阅网上资料,未找到本条文献卷期页码,请确认).