基于改进CLDNN的辐射源信号识别

doi:10.3969/j.issn.1001-506X.2021.01.06

系统工程与电子技术 ›› 2021, Vol. 43 ›› Issue (1): 42-47.doi: 10.3969/j.issn.1001-506X.2021.01.06

基于改进CLDNN的辐射源信号识别

孙艺聪¹(), 田润澜¹(), 王晓峰¹(), 董会旭¹(), 戴普²()

1. 空军航空大学航空作战勤务学院, 吉林长春 130022
2. 空军实验训练基地二区检验所, 陕西咸阳 713800

收稿日期:2020-05-10 出版日期:2020-12-25 发布日期:2020-12-30
作者简介:孙艺聪(1996-),男,硕士研究生,主要研究方向为电子侦察情报的分析与处理。E-mail:sunyc1996@126.com|田润澜(1973-),女,教授,硕士研究生导师,博士,主要研究方向为航空电子侦察情报分析。E-mail:tianrunlan@126.com|王晓峰(1987-),男,讲师,博士,主要研究方向为信号与信号处理。E-mail:wxf870516@126.com|董会旭(1987-),男,讲师,博士,主要研究方向为雷达信号处理。E-mail:me_isdx@163.com|戴普(1986-),男,工程师,本科,主要研究方向为雷达信号处理。E-mail:357862682@qq.com
基金资助:
国家自然科学基金(61571462)

Emitter signal recognition based on improved CLDNN

Yicong SUN¹(), Runlan TIAN¹(), Xiaofeng WANG¹(), Huixu DONG¹(), Pu DAI²()

1. School of Aviation Operations and Services, Aviation University Air Force, Changchun 130022, China
2. Second Zone Inspection Institute of Air Force Experimental Training Base, Xianyang 713800, China

Received:2020-05-10 Online:2020-12-25 Published:2020-12-30

摘要/Abstract

摘要：

传统辐射源信号识别方法往往需要人工提取特征,不仅对专业知识要求较高,而且人为选择的特征不能够保证适用于大多数类型信号的识别,识别精度和识别速度也不能兼顾。针对上述问题,将语音处理领域常用的深度学习模型——卷积长短时深度神经网络(convolutional long short-term deep neural network, CLDNN)引入到辐射源信号的识别中,并将该模型中的长短时记忆层改为双向门控循环单元层。模型的输入为原始时间序列数据,特征提取和分类识别过程均在网络中进行,避免了人工选择特征的不完备性。实验结果表明,所提模型在低信噪比情况下也能够有效识别信号类型,同时与其他模型相比,实现了识别精度和识别速度之间的平衡。

关键词: 辐射源信号识别, 深度学习, 卷积长短时深度神经网络, 时间序列

Abstract:

Traditional methods of emitter signal recognition often need to extract features manually, which not only requires high professional knowledge, but also can not guarantee that the features selected by humans are suitable for the recognition of most types of signals, meanwhile, the recognition accuracy and speed cannot be taken into account. To solve the above problems, convolutional long short-term deep neural network(CLDNN), a deep learning model commonly used in speech processing, is introduced into the recognition of emitter signal, and the long short-term memory (LSTM) layer in this model is changed into bidirectional gated recurrent unit (Bi-GRU) layer. The input of the model is the original time series data, and the processes of feature extraction and classification recognition are carried out in the network to avoid the incompleteness of artificial feature selection. Experimental results show that the proposed model can recognize the signal types effectively at low signal to noise ratio, and a balance between recognition accuracy and recognition speed is achieved compared with other models.

Key words: emitter signal recognition, deep learning, convolutional long short-term deep neural network (CLDNN), time series

中图分类号:

TN971.1

孙艺聪, 田润澜, 王晓峰, 董会旭, 戴普. 基于改进CLDNN的辐射源信号识别[J]. 系统工程与电子技术, 2021, 43(1): 42-47.

Yicong SUN, Runlan TIAN, Xiaofeng WANG, Huixu DONG, Pu DAI. Emitter signal recognition based on improved CLDNN[J]. Systems Engineering and Electronics, 2021, 43(1): 42-47.

图/表 11

图1

图2

图3

图4

表1

表2

图5

图6

图7

图8

表3

参考文献 21

1	李昆, 朱卫纲. 基于机器学习的雷达辐射源识别综述[J]. 电子测量技术, 2019, 42 (18): 69- 75.
	LI K , ZHU W G . A review of radar radiation source recognition based on machine learning[J]. Electronic Measurement Technology, 2019, 42 (18): 69- 75.
2	张政超, 李应升, 王磊, 等. 雷达辐射源信号识别研究综述[J]. 舰船电子工程, 2009, 29 (4): 10- 14.
	ZHANG Z C , LI Y S , WANG L , et al. Research review of radar radiation source signal recognition[J]. Ship Electronic Engineering, 2009, 29 (4): 10- 14.
3	WANG S Q , BAI J , HUANG X Y , et al. Analysis of radar emitter signal sorting and recognition model structure[J]. Procedia Computer Science, 2019, 154, 500- 503.
4	韩立辉, 黄高明, 王鹏. 基于瞬时自相关算法的线性调频雷达信号脉内分析研究[J]. 舰船电子对抗, 2011, 34 (3): 1- 4, 12.
	HAN L H , HUANG G M , WANG P . Intra-pulse analysis of linear FM radar based on instantaneous autocorrelation algorithm[J]. Shipboard Electronic Countermeasure, 2011, 34 (3): 1- 4, 12.
5	吕博群, 沈永健, 周云生. 基于矩特征的雷达信号脉内调制样式识别方法研究[J]. 遥测遥控, 2017, 38 (4): 32- 37.
	LYU B Q , SHEN Y J , ZHOU Y S . Research on the recognition method of pulse modulation in radar signal based on moment characteristics[J]. Telemetry and Remote Control, 2017, 38 (4): 32- 37.
6	郭立民, 陈鑫, 陈涛. 基于AlexNet模型的雷达信号调制类型识别[J]. 吉林大学学报(工学版), 2019, 49 (3): 1000- 1008.
	GUO L M , CHEN X , CHEN T . Radar signal modulation type recognition based on AlexNet model[J]. Journal of Jilin University (Engineering and Technology Edition), 2019, 49 (3): 1000- 1008.
7	ZHANG M , LIU L T , DIAO M . LPI radar waveform recognition based on time-frequency distribution[J]. Sensors, 2016, 16, 1682.
8	LI Y J , XIAO P , WU H C , et al. LPI radar signal detection based on radial integration of Choi-Williams time-frequency image[J]. Journal of Systems Engineering and Electronics, 2015, 26 (5): 973- 981.
9	刘括然. 基于LSTM的雷达辐射源识别技术[J]. 舰船电子工程, 2019, 39 (12): 92- 95.
	LIU K R . Radar radiation source identification based on LSTM[J]. Ship Electronic Engineering, 2019, 39 (12): 92- 95.
10	陈森森.基于RNN的雷达辐射源分类识别算法研究[D].西安:西安电子科技大学, 2019.
	CHEN S S. Research on radar source classification and recognition algorithm based on RNN[D]. Xi'an: Xidian University, 2019.
11	SAINATH T N, VINYALS O, SENIOR A, et al. Convolutional, long short-term memory, fully connected deep neural networks[C]//Proc.of the IEEE International Conference on Acoustics, Speech and Signal, 2015.
12	DINKEL H, CHEN N, QIAN Y, et al. End-to-end spoofing detection with raw waveform CLDNNS[C]//Proc.of the IEEE International Conference on Acoustics, Speech and Signal, 2017: 4860-4864.
13	ZAZO R, SAINATH T N, SIMKO G, et al. Feature learning with raw-waveform cldnns for voice activity detection[C]//Proc.of the Interspeech, 2016: 3668-3672.
14	SAINATH T N, WEISS R, WILSON K W, et al. Learning the speech front-end with raw waveform CLDNNs[C]//Proc.of the Conference of the International Speech Communication Association, 2015.
15	CHUNG J, GULCEHRE C, CHO K, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[EB/OL].[2020-04-30]. http://arxiv.org/pdf/1412.3555v1.
16	SUNDERMEYER M, SCHLUTER R, NEY H. LSTM neural networks for language modeling[C]//Proc.of the Interspeech, 2012: 601-608.
17	IRIE K, TUSKE Z, ALKHOULI T, et al. LSTM, GRU, highway and a bit of attention: an empirical overview for language modeling in speech recognition[C]//Proc.of the Interspeech, 2016.
18	CHO K, VAN M B, GULCEHRE C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[EB/OL].[2020-04-30]. http://arxiv.org/abs/1406.1078.
19	张国豪, 刘波. 采用CNN和Bidirectional GRU的时间序列分类研究[J]. 计算机科学与探索, 2019, 13 (6): 916- 927.
	ZHANG G H , LIU B . Research on time series classification using CNN and Bidirectional GRU[J]. Journal of Frontiers of Computer Science and Technology, 2019, 13 (6): 916- 927.
20	CHEN J X , JIANG D M , ZHANG Y N , et al. A hierarchical bidirectional GRU model with attention for EEG-Based emotion classification[J]. IEEE Access, 2019, 7, 118530- 118540.
21	王静.基于双向门控循环单元的评论文本情感分类[D].南京:南京邮电大学, 2018.
	WANG J. Sentiment classification for review texts via bi-directional gated recurrent unit[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2018.

信号类型	参数	取值
BPSK	巴克码位数	{7, 11, 13}
Costas	频率/kHz	{[4 7 1 6 5 2 3], [2 6 3 8 7 5 1]}
Costas	采样频率/kHz	{15, 17}
FMCW	调制周期/ms	{50, 25, 35}
FMCW	调制带宽/kHz	{0.25, 0.35, 0.5}
Frank	码位数	{36, 64}
P1~P4	码位数	{36, 64}

序号	结构	损失
1	3CNN+1Bi-GRU+3DNN	0.413
2	3CNN+2Bi-GRU+3DNN	0.503
3	4CNN+1Bi-GRU+3DNN	0.432
4	2CNN+1Bi-GRU+3DNN	0.419
5	3CNN+1Bi-GRU+4DNN	0.430
6	3CNN+1Bi-GRU+2DNN	0.420

模型	总用时/s	训练轮数	平均用时/s
本文模型	3 919	86	46
RNN	1 816	100	18
LSTM	13 469	100	135
Bi-LSTM	37 016	97	382
Bi-GRU	22 536	94	240
CLDNN	3 326	100	33
AlexNet	2 254	100	23

[1]	韩啸, 陈世文, 陈蒙, 杨锦程. 基于互易点学习的LPI信号开集识别[J]. 系统工程与电子技术, 2022, 44(9): 2752-2759.
[2]	张立民, 谭凯文, 闫文君, 张聿远. 基于多级跳线残差网络的雷达辐射源识别[J]. 系统工程与电子技术, 2022, 44(7): 2148-2156.
[3]	金国栋, 薛远亮, 谭力宁, 许剑锟. 基于孪生神经网络的目标跟踪算法进展研究[J]. 系统工程与电子技术, 2022, 44(6): 1805-1822.
[4]	赵晓枫, 徐叶斌, 吴飞, 牛家辉, 蔡伟, 张志利. 基于全局感知机制的地面红外目标检测方法[J]. 系统工程与电子技术, 2022, 44(5): 1461-1467.
[5]	邹虹, 白陈阳, 何鹏, 崔亚平, 王汝言, 吴大鹏. 基于分布式深度学习的边缘服务放置策略[J]. 系统工程与电子技术, 2022, 44(5): 1728-1737.
[6]	陈冬, 句彦伟. 基于语义分割实现的SAR图像舰船目标检测[J]. 系统工程与电子技术, 2022, 44(4): 1195-1201.
[7]	孙晶明, 虞盛康, 孙俊. 基于深度学习的HRRP识别姿态敏感性分析[J]. 系统工程与电子技术, 2022, 44(3): 802-807.
[8]	张心宇, 刘源, 宋佳凝. 基于LSTM神经网络的短期轨道预报[J]. 系统工程与电子技术, 2022, 44(3): 939-947.
[9]	孙世岩, 张钢, 梁伟阁, 佘博, 田福庆. 基于时间序列数据扩增和BLSTM的滚动轴承剩余寿命预测方法[J]. 系统工程与电子技术, 2022, 44(3): 1060-1068.
[10]	姚云翔, 陈莹. 注意力机制下双模态交互融合的目标跟踪网络[J]. 系统工程与电子技术, 2022, 44(2): 410-419.
[11]	金涛, 王晓峰, 田润澜, 张歆东. 基于改进1DCNN+TCN的雷达辐射源快速识别方法[J]. 系统工程与电子技术, 2022, 44(2): 463-469.
[12]	徐桂光, 王旭东, 汪飞, 胡国兵, 高涌荇, 罗泽虎. 基于SE-ResNeXt网络的低信噪比LPI雷达辐射源信号识别[J]. 系统工程与电子技术, 2022, 44(12): 3676-3684.
[13]	高涌荇, 王旭东, 汪玲, 朱岱寅, 郭军, 孟凡旺. 基于RCNN的双极化气象雷达天气信号检测[J]. 系统工程与电子技术, 2022, 44(11): 3380-3387.
[14]	周毅恒, 杨军, 夏赛强, 吕明久. 闪烁现象下旋翼目标微动参数估计方法[J]. 系统工程与电子技术, 2022, 44(1): 54-63.
[15]	付蓉, 黄天耀, 刘一民. 基于深度学习的捷变相参雷达1-Bit块稀疏重构[J]. 系统工程与电子技术, 2022, 44(1): 70-75.

基于改进CLDNN的辐射源信号识别

Emitter signal recognition based on improved CLDNN

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 21

相关文章 15

编辑推荐

Metrics

本文评价