中国人民解放军战略支援部队信息工程大学,河南郑州 450001
[ "李 钦 女,1996年11月出生于河南省郑州市.现为战略支援部队信息工程大学研究生.主要研究方向为雷达信号处理. E-mail: liqin_9503@163.com" ]
刘 伟 男,1980年6月出生于江西省萍乡市.现为战略支援部队信息工程大学副教授.主要研究方向为人工智能、雷达信号处理. E-mail: greatliuliu@163.com
牛朝阳 男,1981年10月出生于安徽省阜阳市. 现为战略支援部队信息工程大学副教授.主要研究方向为人工智能、雷达信号处理. E-mail: ncy_100@163.com
宝音图 男,1988年10月出生于吉林省前郭市.现为战略支援部队信息工程大学研究生.主要研究方向为人工智能、遥感图像处理. E-mail: bao258456@163.com
惠周勃 男,1988年8月出生于陕西省周至市.现为战略支援部队信息工程大学研究生.主要研究方向为智能信号处理. E-mail: 463700123@qq.com
收稿:2021-05-22,
修回:2022-04-04,
纸质出版:2023-03-25
移动端阅览
李钦,刘伟,牛朝阳等.低信噪比下基于分裂EfficientNet网络的雷达信号调制方式识别[J].电子学报,2023,51(03):675-686.
LI Qin,LIU Wei,NIU Chao-yang,et al.Radar Signal Modulation Recognition Based on Split EfficientNet Under Low Signal-to-Noise Ratio[J].ACTA ELECTRONICA SINICA,2023,51(03):675-686.
李钦,刘伟,牛朝阳等.低信噪比下基于分裂EfficientNet网络的雷达信号调制方式识别[J].电子学报,2023,51(03):675-686. DOI: 10.12263/DZXB.20210656.
LI Qin,LIU Wei,NIU Chao-yang,et al.Radar Signal Modulation Recognition Based on Split EfficientNet Under Low Signal-to-Noise Ratio[J].ACTA ELECTRONICA SINICA,2023,51(03):675-686. DOI: 10.12263/DZXB.20210656.
针对低信噪比条件下复杂多类雷达信号调制方式识别率低的问题,本文提出了一种基于时频分析和深度学习的雷达信号调制方式识别方法. 利用CTFD(Cohen class Time-Frequency Distribution)时频分析将信号时域波形变换为二维时频图像,更清晰地表征信号特征;采用灰度化和双三次插值运算等方法对时频图预处理,实现图像通道数和尺寸的减少,以降低深度学习模型数据输入量;进一步调整输入输出通道数构建小型EfficientNet网络,再由多个小型网络并行处理构建分裂网络EfficientNet-B0-Split3,将时频图像输入网络实现雷达信号调制方式识别. 实验结果表明,在信噪比为-8 dB时,新方法对17类不同调制方式的雷达信号整体识别率可达97.1%,相对于扩张残差网络提高约2.4个百分点;在信噪比为-10 dB时,识别率可达92.1%,相对于EfficientNet提高约0.7个百分点,提升了低信噪比条件下复杂多类雷达信号调制方式识别率.
Aiming at the low recognition rate of complex and multi-class radar signal modulation under the condition of low signal-to-noise ratio (SNR)
this paper proposes a radar signal modulation recognition based on time-frequency analysis and deep learning. Using Cohen class time-frequency distribution (CTFD)
transform the signal time-domain waveform into a two-dimensional time-frequency image to characterize the signal characteristics more clearly. Using methods such as grayscale and bicubic interpolation
preprocess the time-frequency image to reduce the number and size of image channels and the amount of data input for deep learning models. Further
adjusting the number of input and output channels to build a small EfficientNet. Then multiple small networks are processed in parallel to construct a split network EfficientNet-B0-Split3. The network gets the inputted time-frequency images and realizes the radar signal modulation recognition. The experimental results show that when the SNR is -8 dB
the overall recognition rate of the new method for radar signals with 17 different modulation methods can reach 97.1%
which is about 2.4 percentage points higher than dilated residual network(DRN); when the SNR is -10 dB
the recognition rate can reach 92.1%
which is about 0.7 percentage points higher than EfficientNet
improves the recognition rate of complex and multi-class radar signal modulation methods under the condition of low SNR.
WEI S J , QU Q Z , SU H , et al . Intra-pulse modulation radar signal recognition based on CLDN network [J]. IET Radar , Sonar & Navigation, 2020 , 14 ( 6 ): 803 - 810 .
WU B , YUAN S B , LI P , et al . Radar emitter signal recognition based on one-dimensional convolutional neural network with attention mechanism [J]. Sensors (Basel, Switzerland) , 2020 , 20 ( 21 ): 6350 .
黄颖坤 , 金炜东 , 余志斌 , 等 . 基于深度学习和集成学习的辐射源信号识别 [J]. 系统工程与电子技术 , 2018 , 40 ( 11 ): 2420 - 2425 .
HUANG Y K , JIN W D , YU Z B , et al . Radar emitter signal recognition based on deep learning and ensemble learning [J]. Systems Engineering and Electronics , 2018 , 40 ( 11 ): 2420 - 2425 . (in Chinese)
李东瑾 , 杨瑞娟 , 李晓柏 , 等 . 基于栈式稀疏降噪自编码网络的辐射源调制识别 [J]. 电子学报 , 2020 , 48 ( 6 ): 1198 - 1204 .
LI D J , YANG R J , LI X B , et al . Emitter signal modulation recognition based on stacked sparse denoising auto-encoders [J]. Acta Electronica Sinica , 2020 , 48 ( 6 ): 1198 - 1204 . (in Chinese)
WEI S J , QU Q Z , SU H , et al . Intra-pulse modulation radar signal recognition based on Squeeze-and-Excitation networks [J]. Signal, Image and Video Processing , 2020 , 14 ( 6 ): 1133 - 1141 .
LI J , ZHANG H Q , OU J P , et al . A radar signal recognition approach via IIF-Net deep learning models [J]. Computational Intelligence and Neuroscience , 2020 , 2020. 8858588 .
肖易寒 , 王亮 , 郭玉霞 . 基于去噪卷积神经网络的雷达信号调制类型识别 [J]. 电子与信息学报 , 2021 , 43 ( 8 ): 2300 - 2307 .
XIAO Y H , WANG L , GUO Y X . Radar signal modulation type recognition based on denoising convolutional neural network [J]. Journal of Electronics & Information Technology , 2021 , 43 ( 8 ): 2300 - 2307 . (in Chinese)
张贤达 , 保铮 . 非平稳信号分析与处理 [M]. 北京 : 国防工业出版社 , 1998 .
COHEN L . Time-frequency distributions-a review [J]. Proceedings of the IEEE , 1989 , 77 ( 7 ): 941 - 981 .
COHEN L . Generalized phase-space distribution functions [J]. Journal of Mathematical Physics , 1966 , 7 ( 5 ): 781 - 786 .
QU Z Y , MAO X J , DENG Z A . Radar signal intra-pulse modulation recognition based on convolutional neural network [J]. IEEE Access , 2018 . 6 : 43874 - 43884 .
秦鑫 , 黄洁 , 查雄 , 等 . 基于扩张残差网络的雷达辐射源信号识别 [J]. 电子学报 , 2020 , 48 ( 3 ): 456 - 462 .
QIN X , HUANG J , ZHA X , et al . Radar emitter signal recognition based on dilated residual network [J]. Acta Electronica Sinica , 2020 , 48 ( 3 ): 456 - 462 . (in Chinese)
TAN M X , LE Q . EfficientNet: rethinking model scaling for convolutional neural networks [C]// International Conference on Machine Learning . California : PMLR , 2019 : 6105 - 6114 .
ZHAO Shuai , ZOU Liguang , WANG Wenxiao , et al . SplitNet: divide and co-training [EB/OL]. ( 2020-11-30 ). https://arxiv.org/abs/2011.14660 https://arxiv.org/abs/2011.14660 .
刘锋 , 黄宁 , 王泽众 . 复杂调制信号截获、分选与识别 [M]. 北京 : 国防工业出版社 , 2015 .
刘锋 , 黄宁 , 王泽众 . 复杂信号侦察理论及应用-(上) [M]. 北京 : 科学出版社 , 2016 .
LI D J , YANG R J , LI X B , et al . Radar signal modulation recognition based on deep joint learning [J]. IEEE Access , 2020 , 8 : 48515 - 48528 .
0
浏览量
12
下载量
1
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621