Automatic Modulation Classification Method Based on Deep Learning Network Fusion
-
摘要: 基于深度学习网络的自动调制分类（Automatic Modulation Classification， AMC）方法虽然对大多数通信调制信号能够取得满意的分类效果，但对WBFM（Wide Band Frequency Modulation）信号和MQAM（Multiple Quadrature Amplitude Modulation）信号的分类并不理想。针对WBFM信号误判的问题，使用判决法来筛选WBFM信号；考虑到信号样本不平衡的情况，引入数据增强方法扩充筛选后的WBFM信号。针对MQAM信号混淆的问题，利用分数阶傅里叶变换（Fractional Fourier Transform， FRFT）获取时频维度更多的特征信息。在此基础上，提出一种基于特征金字塔网络和长短时记忆网络并联的多通道特征融合网络（Multi-channel Feature Fusion， MFF）来提取信号的深层特征和浅层特征进行分类。实验结果表明：本文所提方法在一定程度上能够解决WBFM信号的误判问题和MQAM信号的混淆问题；与CNN（Convolutional Neural Network）、ResNet（Residual Network）、LSTM（Long Short Term Memory）、CLDNN（Convolutional Long Short-term Deep Neural Network）网络相比，所提网络具有更高的分类准确率。Abstract: Although the automatic modulation classification （AMC） method based on deep learning network can achieve satisfactory classification results for most communication modulation signals， it is not ideal for the classification of WBFM signals and MQAM signals. Aiming at the problem of misjudgment of WBFM signal， the decision method is used to screen it. Considering the imbalance of signal samples，a data enhancement method is introduced to expand the screened WBFM signals. To solve the problem of confusion of MQAM signals， fractional Fourier transform （FRFT） is adopted to obtain more feature information of time-frequency dimension. On this basis，a multi-channel feature fusion network that is composed of feature pyramid network and long short-term memory network is proposed to extract deep and shallow features for classification. The experiment results show that the proposed method can solve the problem of misjudgment of WBFM signals and the confusion of MQAM signals to a certain extend. Compared with CNN， ResNet， LSTM and CLDNN networks， the proposed network has higher classification accuracy.
-
-
![]()
图 1 处在静音期的WBFM和DSB-AM信号波形
Figure 1. WBFM and DSB-AM signal waveforms during silent period
![]()
图 5 RML2022.HJ的CNN混淆矩阵（SNR=16 dB）
Figure 5. Convolutional neural network confusion matrix for RML2022.HJ（SNR=16 dB）
![]()
图 6 RML2016.10a的CNN混淆矩阵（SNR=16 dB）
Figure 6. Convolutional neural network confusion matrix for RML2016.10a （SNR=16 dB）
![]()
图 7 MFF混淆矩阵（SNR=16 dB）
Figure 7. Multi-channel feature fusion confusion matrix（SNR=16 dB）
![]()
图 9 不同网络模型分类性能比较
Figure 9. Comparison of classification performance of different network models
表 1 不同网络模型的训练时间对比
Table 1 Comparison of training time of different network models
网络模型 每step训练时间/ms MFF 64 CNN 6 LSTM 29 CLDNN 39 ResNet 21 
下载: 导出CSV
-
[1] CAI Tian，WANG Cheng，CUI Gaofeng，et al. Constellation-wavelet transform automatic modulation identifier for M-ary QAM signals［C］// 2015 IEEE 26th Annual International Symposium on Personal，Indoor，and Mobile Radio Communications. Hong Kong，China. IEEE，2015：212- 216. doi：10.1109/pimrc.2015.7343297 doi: 10.1109/pimrc.2015.7343297
[2] ZHANG Haijian，YU Lei，XIA Guisong. Iterative time-frequency filtering of sinusoidal signals with updated frequency estimation［J］. IEEE Signal Processing Letters，2016，23（1）：139- 143. doi：10.1109/lsp.2015.2504565 doi: 10.1109/lsp.2015.2504565
[3] LIU Xueyuan，SHANG Jing，LEONG P H W，et al. Modulation recognition using an FPGA-based convolutional neural network［C］// 2019 22nd International Conference on Electrical Machines and Systems（ICEMS）. Harbin，China. IEEE，2019：1- 6. doi：10.1109/icems.2019.8922403 doi: 10.1109/icems.2019.8922403
[4] FEHSKE A，GAEDDERT J，REED J H. A new approach to signal classification using spectral correlation and neural networks［C］// First IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks，2005. DySPAN 2005. Baltimore，MD，USA. IEEE，2005：144- 150. doi：10.1109/dyspan.2005.1541789 doi: 10.1109/dyspan.2005.1541789
[5] POSTADJIAN T，BRIS A L，MALLET C，et al. Superpixel partitioning of very high resolution satellite images for large-scale classification perspectives with deep convolutional neural networks［C］// IGARSS 2018- 2018 IEEE International Geoscience and Remote Sensing Symposium. Valencia，Spain. IEEE，2018：1328- 1331. doi：10.1109/igarss.2018.8519222 doi: 10.1109/igarss.2018.8519222
[6] 张文宇，刘畅. 卷积神经网络算法在语音识别中的应用［J］. 信息技术，2018，42（10）：147- 152. ZHANG Wenyu，LIU Chang. Application of convolutional neural network algorithm in speech recognition［J］. Information Technology，2018，42（10）：147- 152.（in Chinese）
[7] CHUNG J S，SENIOR A，VINYALS O，et al. Lip reading sentences in the wild［C］// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu，HI，USA. IEEE，2017：3444- 3453. doi：10.1109/cvpr.2017.367 doi: 10.1109/cvpr.2017.367
[8] O’SHEA T J，CORGAN J，CLANCY T C. Convolutional radio modulation recognition networks［C］// Engineering Applications of Neural Networks，2016：213- 226. doi：10.1007/978-3-319-44188-7_16 doi: 10.1007/978-3-319-44188-7_16
[9] LIU Xiaoyu，YANG Diyu，GAMAL A E. Deep neural network architectures for modulation classification［C］// 2017 51st Asilomar Conference on Signals，Systems，and Computers. Pacific Grove，CA，USA. IEEE，2017：915- 919. doi：10.1109/acssc.2017.8335483 doi: 10.1109/acssc.2017.8335483
[10] YAO Tianyao，CHAI Yuan，WANG Shuai，et al. Radio signal automatic modulation classification based on deep learning and expert features［C］// 2020 IEEE 4th Information Technology，Networking，Electronic and Automation Control Conference. Chongqing，China. IEEE，2020，1：1225- 1230. doi：10.1109/itnec48623.2020.9085077 doi: 10.1109/itnec48623.2020.9085077
[11] CHAITHANYA V，REDDY V U. Blind modulation classification in the presence of carrier frequency offset［C］// 2010 International Conference on Signal Processing and Communications（SPCOM）. Bangalore，India. IEEE，2010：1- 5. doi：10.1109/spcom.2010.5560548 doi: 10.1109/spcom.2010.5560548
[12] 孙伟豹，孙虎，王军. 判决法对数字和模拟信号的自动识别及验证［J］. 计算机科学，2012，39（S3）：88- 89，107. doi：10.3969/j.issn.1002-137X.2012.z3.024 doi: 10.3969/j.issn.1002-137X.2012.z3.024 SUN Weibao，SUN Hu，WANG Jun. Automatic recognition of digital and analog communication signals based on decision laws［J］. Computer Science，2012，39（S3）：88- 89，107.（in Chinese）. doi：10.3969/j.issn.1002-137X.2012.z3.024 doi: 10.3969/j.issn.1002-137X.2012.z3.024
[13] 梁艳，梁昔明，廖力清. 模拟信号调制方式自动识别仿真［J］. 计算机测量与控制，2006，14（1）：117- 119，127. doi：10.3969/j.issn.1671-4598.2006.01.043 doi: 10.3969/j.issn.1671-4598.2006.01.043 LIANG Yan，LIANG Ximing，LIAO Liqing. Simulation of automatic recognition of analogue modulated signals［J］. Computer Measurement＆ Control，2006，14（1）：117- 119，127.（in Chinese）. doi：10.3969/j.issn.1671-4598.2006.01.043 doi: 10.3969/j.issn.1671-4598.2006.01.043
[14] 甘俊英，翟懿奎，黄聿，等. 基于双激活层深度卷积特征的人脸美丽预测研究［J］. 电子学报，2019，47（3）：636- 642. doi：10.3969/j.issn.0372-2112.2019.03.017 doi: 10.3969/j.issn.0372-2112.2019.03.017 GAN Junying，ZHAI Yikui，HUANG Yu，et al. Research of facial beauty prediction based on deep convolutional features using double activation layer［J］. Acta Electronica Sinica，2019，47（3）：636- 642.（in Chinese）. doi：10.3969/j.issn.0372-2112.2019.03.017 doi: 10.3969/j.issn.0372-2112.2019.03.017
[15] 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 doi: 10.1109/tccn.2018.2835460
[16] 张忠民，刘刚，刘鲁涛. 基于分数阶傅里叶变换和循环谱的雷达信号调制方式识别［J］. 应用科技，2020，47（3）：30- 36. doi：10.11991/yykj.201909013 doi: 10.11991/yykj.201909013 ZHANG Zhongmin，LIU Gang，LIU Lutao. Radar signal modulation recognition based on fractional Fourier transform and cyclic spectrum［J］. Applied Science and Technology，2020，47（3）：30- 36.（in Chinese）. doi：10.11991/yykj.201909013 doi: 10.11991/yykj.201909013
[17] WEST N E，O'SHEA T. Deep architectures for modulation recognition［C］// 2017 IEEE International Symposium on Dynamic Spectrum Access Networks（DySPAN）. Baltimore，MD，USA. IEEE，2017：1- 6. doi：10.1109/dyspan.2017.7920754 doi: 10.1109/dyspan.2017.7920754
[18] LIN T Y，DOLLÁR P，GIRSHICK R，et al. Feature pyramid networks for object detection［C］// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu，HI，USA. IEEE，2017：936- 944. doi：10.1109/cvpr.2017.106 doi: 10.1109/cvpr.2017.106
[19] EROL B，LEE D S，HULL J. Multimodal summarization of meeting recordings［C］// 2003 International Conference on Multimedia and Expo. ICME' 03. Proceedings（Cat. No.03TH8698）. Baltimore，MD，USA. IEEE，2003：III-25. doi：10.1109/icme.2003.1221239 doi: 10.1109/icme.2003.1221239
[20] HONG Dehua，ZHANG Zilong，XU Xiaodong. Automatic modulation classification using recurrent neural networks［C］// 2017 3rd IEEE International Conference on Computer and Communications. Chengdu，China. IEEE，2017：695- 700. doi：10.1109/compcomm.2017.8322633 doi: 10.1109/compcomm.2017.8322633
-
期刊类型引用(2)
1. 丁功鸿，黄山. 结合多特征与线性判别分析的图像检索. 计算机应用与软件. 2024(04): 212-218 . 
百度学术
2. 李薇. 基于云平台和分布式计算的大规模图像检索. 微型电脑应用. 2024(08): 211-215 . 百度学术
其他类型引用(0)
计量
- 文章访问数: 176
- HTML全文浏览量: 62
- PDF下载量: 72
- 被引次数: 2
