基于集成深度学习的有源干扰智能分类

doi:10.12305/j.issn.1001-506X.2022.12.02

摘要/Abstract

摘要：

针对现有基于机器学习的雷达有源干扰分类大多需要构建人工特征集且小样本情况下分类精度低的问题, 提出一种基于多通道特征融合的集成卷积神经网络(convolutional neural network, CNN)分类方法。首先, 建立多种有源干扰的数学模型, 仿真并利用短时傅里叶变换获得其时频分布图; 其次, 提取时频分布图的实部、虚部和模值三通道特征, 通过多种特征组合方式建立不同特征组合的样本集; 最终, 构建以CNN为基分类器的集成深度学习模型, 每个CNN分别提取不同样本集的特征, 对所有基分类器的预测结果做多数投票得到集成模型的整体预测结果。实验表明, 该方法能够有效实现小样本情况下多类有源干扰的高精度智能化识别。

关键词: 有源干扰分类, 短时傅里叶变换, 集成学习, 卷积神经网络, 小样本

Abstract:

Aiming at the problems that most existing machine learning-based classification of radar active jamming need to construct artificial feature sets and low classification accuracy in the case of small samples, an ensemble convolutional neural network (CNN) classification method based on multichannel feature fusion is proposed. Firstly, multiple mathematical models of active jamming are established, simulated and the corresponding time-frequency profiles are obtained by using the short-time Fourier transform (STFT). Secondly, real part, imaginary part, and modulus three-channel features of time-frequency distribution plots are extracted to establish sample sets containing different combinations of features through multiple feature combinations. Ultimately, an ensemble depth model with CNN as the base classifier is constructed, each CNN separately extracts the features of different sample sets, and majority voting on the prediction results of all base classifiers gives the overall prediction results of the ensemble model. The experiments show that the proposed method can effectively realize highly accurate intelligent identification of multiple classes of active jamming in the case of small samples.

Key words: active jamming classification, short-time Fourier transform (STFT), ensemble learning, convolutional neural network (CNN), small sample

中图分类号:

TN974

吕勤哲, 全英汇, 沙明辉, 董淑仙, 邢孟道. 基于集成深度学习的有源干扰智能分类[J]. 系统工程与电子技术, 2022, 44(12): 3595-3602.

Qinzhe LYU, Yinghui QUAN, Minghui SHA, Shuxian DONG, Mengdao XING. Ensemble deep learning-based intelligent classification of active jamming[J]. Systems Engineering and Electronics, 2022, 44(12): 3595-3602.

图/表 8

图1

表1

图2

图3

表2

表3

图4

表4

参考文献 30

1	张光义. 提高雷达系统抗干扰能力的一些措施[J]. 现代雷达, 2001, 23 (1): 6- 12. doi: 10.3969/j.issn.1004-7859.2001.01.002
	ZHANG G Y . Some measures for enhancing ECCM capabilities of radar systems[J]. Modern Radar, 2001, 23 (1): 6- 12. doi: 10.3969/j.issn.1004-7859.2001.01.002
2	TAN M , WANG C Y , XUE B , et al. A novel deceptive jamming approach against frequency diverse array radar[J]. IEEE Sensors Journal, 2021, 21 (6): 8323- 8332. doi: 10.1109/JSEN.2020.3045757
3	OLIVIER K , CILLIERS J E , PLESSIS M D . Design and performance of wideband DRFM for radar test and evaluation[J]. Electronics Letters, 2011, 47 (14): 824- 825. doi: 10.1049/el.2011.0362
4	WANG Y F , SUN B Y , WANG N . Recognition of radar active-jamming through convolutional neural networks[J]. The Journal of Engineering, 2019, 2019 (21): 7695- 7697. doi: 10.1049/joe.2019.0659
5	张建中, 穆贺强, 文树梁, 等. 基于脉内步进LFM波形的抗间歇采样转发干扰方法[J]. 系统工程与电子技术, 2019, 41 (5): 1013- 1020.
	ZHANG J Z , MU H Q , WEN S L , et al. Anti interrupted-sampling repeater jamming method based on stepped LFM waveform[J]. Systems Engineering and Electronics, 2019, 41 (5): 1013- 1020.
6	SHAO G Q , CHEN Y S , WEI Y S . Deep fusion for radar jamming signal classification based on CNN[J]. IEEE Access, 2020, 8, 117236- 117244. doi: 10.1109/ACCESS.2020.3004188
7	祝存海. 基于特征提取的雷达有源干扰信号分类研究[D]. 西安: 西安电子科技大学, 2017.
	ZHU C H. Research on radar active jamming signal disturbance classification base on feature extraction[D]. Xi'an: Xidian University, 2017.
8	YAN L J , ADDABBO P , ZHANG Y X , et al. A sparse learning approach to the detection of multiple noise-like jammers[J]. IEEE Trans.on Aerospace and Electronic Systems, 2020, 56 (6): 4367- 4383. doi: 10.1109/TAES.2020.2988960
9	GRECO M , GINI F , FARINA A . Radar detection and classification of jamming signals belonging to a cone class[J]. IEEE Trans.on Signal Processing, 2008, 56 (5): 1984- 1993. doi: 10.1109/TSP.2007.909326
10	YANG X Y, RUAN H L, FENG H R. A recognition algorithm of deception jamming based on image of time-frequency distribution[C]//Proc. of the 7th IEEE International Confe-rence on Electronics Information and Emergency Communication, 2017: 275-278.
11	刘明骞, 高晓腾, 张俊林. 多类型的雷达有源干扰感知新方法[J]. 西安交通大学学报, 2019, 53 (10): 103- 108.
	LIU M Q , GAO X T , ZHANG J L . A novel sensing method for multi-types of radar active jamming[J]. Journal of Xi'an Jiaotong University, 2019, 53 (10): 103- 108.
12	WEI D X , ZHANG S N , CHEN S , et al. Research on anti-jamming technology of chaotic composite short range detection system based on underdetermined signal separation and spectral analysis[J]. IEEE Access, 2019, 7, 42298- 42308. doi: 10.1109/ACCESS.2019.2907621
13	LIU Q, ZHANG W. Deep learning and recognition of radar jamming based on CNN[C]//Proc. of the 12th IEEE International Symposium on Computational Intelligence and Design, 2019: 208-212.
14	QIN Y, YANG J, ZHU M T, et al. Fast recognition of pull-off jamming using LSTM[C]//Proc. of the IEEE International Conference on Signal, Information and Data Processing, 2019.
15	刘强. 基于深度学习的雷达干扰识别技术[D]. 成都: 电子科技大学, 2020.
	LIU Q. Radar interference recognition based on deep learning[D]. Chengdu: University of Electronic Science and Technology of China, 2020.
16	刘建洋. 基于时频域分析的电子干扰识别方法研究[D]. 成都: 电子科技大学, 2018.
	LIU J Y. Research on electronic jamming identification method based on time frequency domain analysis[D]. Chengdu: University of Electronic Science and Technology of China, 2018.
17	LIU R H, DUAN K Q. Barrage noise jamming suppression methods for airborne phased array radar[C]//Proc. of the IET International Radar Conference, 2009.
18	郑苡榕. 雷达有源干扰识别方法及高效实现[D]. 成都: 电子科技大学, 2019.
	ZHENG Y R. Radar active jamming recognition methods and efficient implementation[D]. Chengdu: University of Electronic Science and Technology of China, 2019.
19	ZHANG J, WU G, YAN X W. Comparative analysis of sweeping frequency jamming and non-sweeping frequency jamming in partial-band jamming based on projectile-carried communication jamming[C]//Proc. of the IEEE 20th International Conference on Communication Technology, 2020: 379-383.
20	杜聪炬. 雷达非常规有源压制与欺骗干扰智能识别理论与方法研究[D]. 成都: 电子科技大学, 2019.
	DU C J. Study on theory and method of radar unconventional active barrage and deception jamming intelligent recognition[D]. Chengdu: University of Electronic Science and Technology of China, 2019.
21	WEN C , PENG J Y , ZHOU Y , et al. Enhanced three-dimensional joint domain localized STAP for airborne FDA-MIMO radar under dense false-target jamming scenario[J]. IEEE Sensors Journal, 2018, 18 (10): 4154- 4166. doi: 10.1109/JSEN.2018.2820905
22	FENG D J , XU L T , PAN X Y , et al. Jamming wideband radar using interrupted-sampling repeater[J]. IEEE Trans.on Aerospace and Electronic Systems, 2017, 53 (3): 1341- 1354. doi: 10.1109/TAES.2017.2670958
23	HAN B W , YANG X P , WU X C , et al. Smart noise jamming suppression method based on fast fractional filtering[J]. The Journal of Engineering, 2019, 2019 (19): 6201- 6205. doi: 10.1049/joe.2019.0270
24	张磊, 佘忱. 基于随机跳频的末制导雷达抗瞄准式干扰性能研究[J]. 舰船电子对抗, 2020, 43 (5): 17- 19.
	ZHANG L , SHE C . Research into the anti-spot-jamming performance of terminal guidance radar based on random frequency hopping[J]. Shipboard Electronic Countermeasure, 2020, 43 (5): 17- 19.
25	周新, 姚富强, 牛英滔. 一种线性扫频干扰信号的参数估计方法[J]. 通信技术, 2016, 49 (12): 1582- 1587. doi: 10.3969/j.issn.1002-0802.2016.12.002
	ZHOU X , YAO F Q , NIU Y T . Parameter estimation of linear chirp signal[J]. Communications Technology, 2016, 49 (12): 1582- 1587. doi: 10.3969/j.issn.1002-0802.2016.12.002
26	王程, 郑小燕, 王海彬. 基于短时傅里叶变换的干扰信号识别方法[J]. 装备环境工程, 2018, 15 (3): 67- 70.
	WANG C , ZHENG X Y , WANG H B . Interference signal recognition method based on short-time Fourier transform[J]. Equipment Environmental Engineering, 2018, 15 (3): 67- 70.
27	HLAWATSCH F , BOUDREAUX-BARTELS G F . Linear and quadratic time-frequency signal representations[J]. IEEE Signal Processing Magazine, 1992, 9 (2): 21- 67. doi: 10.1109/79.127284
28	晁娇. 基于短时傅里叶变换的雷达信号脉内特征自动识别研究[J]. 现代信息科技, 2019, 3 (1): 50- 53.
	CHAO J . An auto recognition method for in-pulse features of radar signal based on STFT[J]. Modern Information Technology, 2019, 3 (1): 50- 53.
29	TIN K H . The random subspace method for constructing decision forests[J]. IEEE Trans.on Pattern Analysis & Machine Intelligence, 1998, 20 (8): 832- 844.
30	LI S T , SONG W W , FANG L Y , et al. Deep learning for hyperspectral image classification: an overview[J]. IEEE Trans.on Geoscience and Remote Sensing, 2019, 57 (9): 6690- 6709.

干扰类型	参数类型	取值范围
DFTJ	假目标数/个	3~6
DFTJ	假目标延迟/μs	1~10
ISRJ+SNJ	转发次数/次	1~4
	干扰占空比/%	20~50
	采样周期/μs	5, 10
RDJ	假目标延迟/μs	1~10
SJ	干扰带宽/MHz	20~40
BJ	干扰带宽/MHz	50~80
LSJ	扫频周期μs	40~80
LSJ	扫频带宽/MHz	20

网络层	卷积核	最大池化	激活函数
输入层		N×(247×100×3)
卷积层1	(13×13)×64	2×2	ReLU
正则化层		Dropout(0.25)
卷积层2	(9×9)×128	2×2	ReLU
卷积层3	(5×5)×128	2×2	ReLU
正则化层		Dropout(0.25)
卷积层4	(3×3)×256	2×2	ReLU
Flatten层		1×5 632
全连接层	-	-	Softmax
输出层		1×12

网络层	卷积核	最大池化	激活函数
输入层		N×(2 000×3)
卷积层1	(16×1)×16	2×1	ReLU
正则化层		Dropout(0.25)
卷积层2	(32×1)×32	2×1	ReLU
卷积层3	(64×1)×64	2×1	ReLU
正则化层		Dropout(0.25)
卷积层4	(128×1)×128	2×1	ReLU
Flatten层		1×5 120
全连接层	-	-	Softmax
输出层		1×12

干扰	算法
干扰	时域 CNN	人工特征提取+RF	时频域 CNN	所提 ECNN
TG	88.55±8.80	94.11±3.21	97.32±4.50	99.75±0.24
DFTJ	92.08±3.29	86.10±4.36	92.77±4.34	98.72±0.59
ISRJ	78.15±10.62	98.50±3.47	95.43±3.39	93.46±1.12
RDJ	76.86±13.34	97.32±3.78	95.80±2.00	98.59±0.21
SNJ	76.46±9.86	92.49±2.61	94.89±1.18	83.61±1.32
SJ	99.50±0.74	99.79±0.25	98.24±2.15	99.14±0.55
BJ	92.30±14.52	88.18±5.37	96.33±3.19	100.00±0.00
LSJ	70.47±9.85	87.31±3.82	83.92±7.56	89.23±1.57
RDJ+SNJ	78.95±12.70	90.32±1.65	86.54±7.19	90.77±1.34
DFTJ+SNJ	97.66±2.22	89.55±4.76	96.14±3.19	100.00±0.00
RDJ+ISRJ	83.40±8.96	95.57±1.99	94.50±3.62	96.73±0.91
LSJ+ISRJ	66.04±7.96	89.91±4.54	85.32±10.33	92.90±4.72

[1]	宋爽, 张悦, 张琳娜, 岑翼刚, 李浥东. 基于深度学习的轻量化目标检测算法[J]. 系统工程与电子技术, 2022, 44(9): 2716-2725.
[2]	王彩云, 吴钇达, 王佳宁, 马璐, 赵焕玥. 基于改进的CNN和数据增强的SAR目标识别[J]. 系统工程与电子技术, 2022, 44(8): 2483-2487.
[3]	孙晶明, 虞盛康, 孙俊. 基于元学习的雷达小样本目标识别方法及改进[J]. 系统工程与电子技术, 2022, 44(6): 1839-1845.
[4]	秦博伟, 蒋磊, 许华, 齐子森. 基于残差生成对抗网络的调制识别算法[J]. 系统工程与电子技术, 2022, 44(6): 2019-2026.
[5]	韦娟, 杨皇卫, 宁方立. 基于NMF与CNN联合优化的声学场景分类[J]. 系统工程与电子技术, 2022, 44(5): 1433-1438.
[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]	方伟, 王玉, 闫文君, 林冲. 基于神经网络的符号化飞行动作识别[J]. 系统工程与电子技术, 2022, 44(3): 737-745.
[8]	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.
[9]	刘恒燕, 张立民, 闫文君, 钟兆根, 凌青, 梁晓军. 基于WBP-CNN算法的LDPC译码[J]. 系统工程与电子技术, 2022, 44(3): 1030-1035.
[10]	邵凯, 朱苗苗, 王光宇. 基于生成对抗与卷积神经网络的调制识别方法[J]. 系统工程与电子技术, 2022, 44(3): 1036-1043.
[11]	张玺, 金正猛, 姜亚琴. 融合深度图像先验的全变差图像着色算法[J]. 系统工程与电子技术, 2022, 44(2): 385-393.
[12]	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.
[13]	刘旗, 张新禹, 刘永祥. 基于门控多尺度匹配网络的小样本SAR目标识别[J]. 系统工程与电子技术, 2022, 44(11): 3346-3356.
[14]	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.
[15]	李永刚, 朱卫纲, 黄琼男, 李云涛, 何永华. 复杂背景下SAR图像近岸舰船目标检测[J]. 系统工程与电子技术, 2022, 44(10): 3096-3103.