基于注意力机制和改进CLDNN的雷达辐射源识别

doi:10.12305/j.issn.1001-506X.2021.05.09

系统工程与电子技术 ›› 2021, Vol. 43 ›› Issue (5): 1224-1231.doi: 10.12305/j.issn.1001-506X.2021.05.09

基于注意力机制和改进CLDNN的雷达辐射源识别

崔邦彦¹(), 田润澜^1,*(), 王东风²(), 崔钢¹(), 石静苑³()

1. 空军航空大学航空作战勤务学院, 吉林长春 130022
2. 空军研究院, 北京 100012
3. 空军航空大学航空基础学院, 吉林长春 130022

收稿日期:2020-09-07 出版日期:2021-05-01 发布日期:2021-04-27
通讯作者: 田润澜 E-mail:cby0124@126.com;tianrunlan@126.com;dongfeng_wang@sina.com;cghomer@sina.com;495017115@qq.com
作者简介:崔邦彦(1997—), 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: cby0124@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|王东风(1979—), 男, 高级工程师, 博士, 主要研究方向为电子对抗。E-mail: dongfeng_wang@sina.com|崔钢(1978—), 男, 副教授, 硕士, 主要研究方向为虚拟仿真。E-mail: cghomer@sina.com|石静苑(1973—), 女, 副教授, 硕士, 主要研究方向为控制理论与控制工程。E-mail: 495017115@qq.com
基金资助:
国家自然科学基金(61571462)

Radar emitter identification based on attention mechanism and improved CLDNN

Bangyan CUI¹(), Runlan TIAN^1,*(), Dongfeng WANG²(), Gang CUI¹(), Jingyuan SHI³()

1. School of Aviation Operations and Services, Aviation University of Air Force, Changchun 130022, China
2. Air Force Research Institute, Beijing 100012, China
3. School of Aeronautical Foundation, Aviation University of Air Force, Changchun 130022, China

Received:2020-09-07 Online:2021-05-01 Published:2021-04-27
Contact: Runlan TIAN E-mail:cby0124@126.com;tianrunlan@126.com;dongfeng_wang@sina.com;cghomer@sina.com;495017115@qq.com

摘要/Abstract

摘要：

传统的辐射源识别通过比对、匹配辐射源信号与雷达数据库来识别, 这种方法很难满足战时高效、快速和准确的识别要求。随着机器学习方法的提出, 诸如支持向量机等算法在辐射源识别领域的运用, 可以满足战时高效、快速的识别要求, 但这种方法在低信噪比环境下, 辐射源识别准确率低。针对上述问题, 采用深度学习, 引入注意力机制和特征融合方法, 提出注意力机制特征融合一维卷积长短时深度神经网络(attention-mechanism feature-fusion one-dimensional convolution long-short-term-memory deep neural networks, AF1CLDNN)识别模型。实验验证了注意力机制和特征融合方法的有效性, 及新识别模型在低信噪比环境下具有较高识别准确率与识别速度。

关键词: 辐射源识别, 深度学习, 时间序列, 注意力机制, 特征融合, 一维卷积长短时深度神经网络

Abstract:

Traditional emitter identification is based on the comparison and matching of emitter signal and radar database, which is difficult to meet the requirements of high efficiency, fast and accurate identification in wartime. With the development of machine learning methods, such as the application of support vector machine (SVM) and other algorithm in the field of emitter identification, can meet the requirements of efficient and rapid identification in wartime. However, this method has low accuracy of emitter identification in low signal to noise ratio environment. In order to solve the above problems, the deep learning is used, the attention mechanism and feature fusion method is introduced, and a indentification model of attention-mechanism feature-fusion one-dimensional convolution long-short-term-memory deep neural networks (AF1CLDNN) is proposed. The effectiveness of attention mechanism and feature fusion method is verified by experiments, and the new indentification model has high indentification accuracy and indentification speed in low signal to noise ratio environment.

Key words: emitter identification, deep learning, time series, attention mechanism, feature fusion, one-dimensional convolutional long-short-term-memory deep neural networks(1CLDNN)

中图分类号:

TN971.1

崔邦彦, 田润澜, 王东风, 崔钢, 石静苑. 基于注意力机制和改进CLDNN的雷达辐射源识别[J]. 系统工程与电子技术, 2021, 43(5): 1224-1231.

Bangyan CUI, Runlan TIAN, Dongfeng WANG, Gang CUI, Jingyuan SHI. Radar emitter identification based on attention mechanism and improved CLDNN[J]. Systems Engineering and Electronics, 2021, 43(5): 1224-1231.

图/表 12

图1

图2

图3

图4

表1

表2

表3

图5

表4

图6

图7

图8

参考文献 34

1	WANG Q S , 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. doi: 10.1016/j.procs.2019.06.076
2	孙艺聪, 田润澜, 王晓峰, 等. 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.
	SUN Y C , TIAN R L , WANG X F , et al. Radiation source signal recognition based on improved CLDNN[J]. Systems Engineering and Electronics, 2021, 43 (1): 42- 47.
3	朱志宇, 张冰, 王建华. 灰关联分析和证据理论在雷达辐射源识别中的应用及改进[J]. 电光与控制, 2007, 14 (1): 34- 37. doi: 10.3969/j.issn.1671-637X.2007.01.010
	ZHU Z Y , ZHANG B , WANG J H . Application and improvement of Grey correlation analysis and evidence theory in radar source identification[J]. Electronics Optics and Control, 2007, 14 (1): 34- 37. doi: 10.3969/j.issn.1671-637X.2007.01.010
4	朱志宇. 基于模糊推理的雷达辐射源识别方法[J]. 火力与指挥控制, 2009, 34 (4): 95- 99. doi: 10.3969/j.issn.1002-0640.2009.04.027
	ZHU Z Y . A method of radar source identification based on fuzzy inference[J]. Fire Control and Command Control, 2009, 34 (4): 95- 99. doi: 10.3969/j.issn.1002-0640.2009.04.027
5	WANG Y H , ZHANG S C , ZHANG Y W , et al. A cooperative spectrum sensing method based on empirical mode decomposition and information geometry in complex electromagnetic environment[J]. Complexity, 2019, 2019, 5470974.
6	杜盈, 何瑞珠. 基于深度学习的电磁辐射源识别技术[J]. 通讯世界, 2020, 27 (7): 75- 76.
	DU Y , HE R Z . Electromagnetic radiation source identification technology based on deep learning[J]. Telecom World, 2020, 27 (7): 75- 76.
7	李昆, 朱卫纲. 一种深度学习的雷达辐射源识别方法[J]. 电子设计工程, 2020, 28 (12): 99- 104.
	LI K , ZHU W G . A deep learning radar radiation source identification method[J]. Electronic Design Engineering, 2020, 28 (12): 99- 104.
8	BUKHARI D , WANG Y T , WANG H . Multilingual convolutional, long short-term memory, deep neural networks for low resource speech recognition[J]. Procedia Computer Science, 2017, 107, 842- 847. doi: 10.1016/j.procs.2017.03.179
9	MIAO Y L, JI Y C, PENG E L. Application of CNN-BiGRU model in chinese short text sentiment analysis[C]//Proc. of the International Association of Applied Science and Engineering, 2019: 510-514.
10	LIU Z D , ZHOU W G , LI H Q . AB-LSTM: attention-based bidrectional LSTM model for scene text detection[J]. ACM Trans.on Multimedia Computing Communications, and Applications, 2019, 15 (4): 107.
11	SAINATH T N, WEISS R, WILSON K W. Learning the speech front end with raw waveform CLDNNs[C]//Proc. of the Conference of the International Speech Communication Association, 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 Processing, 2017: 4860-4864.
13	VISHAL P , RAJESH K A . Convolutional support vector machines for speech recogniton[J]. International Journal of Speech Technology, 2019, 22 (3): 601- 609. doi: 10.1007/s10772-018-09584-4
14	VIMINA E R , JACOB K P . Feature fusion method using BoVW framework for enhancing image retrieval[J]. IET Image Processing, 2019, 13 (11): 1979- 1985. doi: 10.1049/iet-ipr.2018.5381
15	HOU S D , SUN Q S . An orthogonal regularized CCA learning algorithm for feature fusion[J]. Journal of Visual Communication and Image Representation, 2014, 25 (5): 785- 792. doi: 10.1016/j.jvcir.2014.01.009
16	LI J , JIN K , ZHOU D L , et al. Attention mechanism-based CNN for facial expression recognition[J]. Neurocomputing, 2020, 411, 340- 350. doi: 10.1016/j.neucom.2020.06.014
17	XU H F , CHAI L , LUO Z M , et al. Stock movement predictive network via incorporative attention mechanisms based on tweet and historical prices[J]. Neurocomputing, 2020, 418, 326- 339. doi: 10.1016/j.neucom.2020.07.108
18	LEI Y T , DU W W , HU Q H . Face sketch-to-photo transformation with multi-scale self-attention GAN[J]. Neurocomputing, 2020, 396, 13- 23. doi: 10.1016/j.neucom.2020.02.024
19	GAO L L , WANG X H , SONG J K , et al. Fused GRU with semantic-temporal attention for video captioning[J]. Neurocomputing, 2020, 395, 222- 228. doi: 10.1016/j.neucom.2018.06.096
20	XU X , WANG W , WANG J H . A three-way incremental-learning algorithm for radar emitter identification[J]. Frontiers of Computer Science, 2016, 10 (4): 673- 688. doi: 10.1007/s11704-015-4457-7
21	CHEN C X , HE M H , LI H F . An improved radar emitter recognition method based on dezert-smarandache theory[J]. Chinese Journal of Electronics, 2015, 24 (3): 611- 615. doi: 10.1049/cje.2015.07.029
22	ZHANG W B , JI H B , LIAO G S , et al. A novel extreme learning machine using privileged information[J]. Neurocomputing, 2015, 168, 823- 828. doi: 10.1016/j.neucom.2015.05.042
23	LI Y B , GE J , LIN Y , et al. Radar emitter signal recognition based on multi-scale wavelet entropy and feature weighting[J]. Journal of Central South University, 2014, 21 (11): 4254- 4260. doi: 10.1007/s11771-014-2422-5
24	CONG C , ZHANG H M . Invert-U-Net DNN segmentation model for MRI cardiac left ventricle segmentation[J]. The Journal of Engineering, 2018, 2018 (16): 1463- 1467. doi: 10.1049/joe.2018.8302
25	LIU M , JIANG H J , HU C . Aperiodically intermittent strategy for finite-time synchronization of delayed neural networks[J]. Neurocomputing, 2018, 310, 1- 9. doi: 10.1016/j.neucom.2018.04.009
26	WEI R , ZHANG X H , WANG J H , et al. The research of sleep staging based on single-lead electrocardiogram and deep neural network[J]. Biomedical Engineering Letters, 2018, 8, 87- 93.
27	ZHANG Y J , QIN N , HUANG D Q , et al. Fault diagnosis of high-speed train bogie based on deep neural network[J]. IFAC PapersOnLine, 2019, 52 (24): 135- 139.
28	GAO Z W , WANG J , WANG J Y , et al. Time-frequency analysis of the vortex motion in a cylindrical cyclone separator[J]. Chemical Engineering Journal, 2019, 373, 1120- 1131.
29	YAO F , CHEN G Y . Time-frequency analysis of impact echo signals of grouting defects in tunnels[J]. Russian Journal of Nondestructive Testing, 2019, 55 (8): 581- 595. doi: 10.1134/S1061830919080102
30	YU G , WANG Z H , ZHAO P , et al. Local maximum synchrosqueezing transform: an energy-concentrated time-frequency analysis tool[J]. Mechanical Systems and Signal Processing, 2019, 117, 537- 552.
31	FU W H , HU Z , LI D . A sorting algorithm for multiple frequency hopping signals in complex electromagnetic environments[J]. Circuits, Systems, and Signal Processing, 2020, 39 (1): 245- 267.
32	ZHANG C H , HAN Y T , ZHANG P , et al. Research on modern radar emitter modelling technique under complex electromagnetic environment[J]. The Journal of Engineering, 2019, 2019 (20): 7134- 7138.
33	LIU C T , WU R J , HE Z X , et al. Modeling and analyzing interference signal in a complex electromagnetic environment[J]. EURASIP Journal on Wireless Communications and Networking, 2016, 2016 (1): 1- 9. doi: 10.1186/s13638-015-0498-8
34	GUAN L , LI Z , HAO B J , et al. Cognitive frequency hopping sequences[J]. Chinese Journal of Electronics, 2016, 25 (1): 185- 191.

信号类型	载频/ kHz	采样频率/ kHz	主要参数
BPSK	1~1.2	7	7, 11, 13位巴克码
Costas	1~1.2	{15, 17}	—
Frank	1~1.2	7	36, 64位码
P1~P4	1~1.2	7	36, 64位码

序号	网络	有无特征融合	训练轮数	训练时间	损失	准确率/%
1	F1CLDNN	有	15	0:16:27	0.130 1	96.02
2	1CLDNN	无	62	5:30:36	0.366 5	86.47

序号	网络	有无注意力	训练轮数	训练时间	损失	准确率/%
1	F1CLDNN	有	10	0:03:46	0.118 2	96.47
2	F1CLDNN	无	15	0:16:27	0.130 1	96.02
3	1CLDNN	有	10	0:55:12	0.291 1	90.99
4	1CLDNN	无	62	5:30:36	0.366 5	86.47

序号	网络	训练时间	训练轮数	损失	准确率/%
1	AF1CLDNN	0:03:46	10	0.118 2	96.47
2	F1CLDNN	0:16:27	15	0.130 1	96.02
3	A1CLDNN	0:55:12	10	0.291 1	90.99
4	1CLDNN	5:30:36	62	0.366 5	86.47
5	DNN	0:05:08	50	0.119 4	96.00
6	Resnet34	0:07:08	7	0.193 3	94.80
7	GoogleNet	0:14:44	13	0.343 7	88.02
8	AlexNet	0:05:55	5	0.261 1	93.39

[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]	赵晓枫, 吴飞, 徐叶斌, 牛家辉, 蔡伟, 张志利. 基于背景还原的红外伪装效果评价方法[J]. 系统工程与电子技术, 2022, 44(8): 2554-2561.
[3]	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.
[4]	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.
[5]	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.
[6]	徐平亮, 崔亚奇, 熊伟, 熊振宇, 顾祥岐. 生成式中断航迹接续关联方法[J]. 系统工程与电子技术, 2022, 44(5): 1543-1552.
[7]	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.
[8]	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.
[9]	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.
[10]	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.
[11]	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.
[12]	吴涛, 王伦文, 朱敬成. 基于迁移学习和注意力机制的伪装图像分割[J]. 系统工程与电子技术, 2022, 44(2): 376-384.
[13]	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.
[14]	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.
[15]	马钰棠, 孙鹏, 张杰勇, 王鹏, 闫云飞, 赵亮. 样本不平衡下的空中群组意图识别方法[J]. 系统工程与电子技术, 2022, 44(12): 3747-3755.

基于注意力机制和改进CLDNN的雷达辐射源识别

Radar emitter identification based on attention mechanism and improved CLDNN

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 34

相关文章 15

编辑推荐

Metrics

本文评价