基于数据流自监督的雷达信号在线分选方法

doi:10.13229/j.cnki.jdxbgxb.20240827

摘要/Abstract

摘要：

针对电子对抗领域雷达信号截获困难和分选准确率低的问题，提出一种基于数据流自监督的雷达信号在线分选新方法。先通过生成对抗网络的均衡特征融合方法，利用有限样本完成高质量特征融合和样本增强；再构建一种基于多任务元学习的自监督模型，实现无标签信号流的在线分选。实验表明，在4 dB时，采用所提方法在仿真数据集上准确率为95.82%，在实测数据和公开数据集上也具有良好表现，证实了所提方法在数据流在线分选方面的有效性。

关键词: 雷达信号, 在线分选, 无监督学习, 数据流

Abstract:

To address the challenges of radar signal interception and low sorting accuracy in the electronic warfare domain， this paper proposes a new online sorting method for radar signals based on self-supervised data streams. Firstly， a balanced feature fusion method using generative adversarial networks is employed to achieve high-quality feature fusion and sample enhancement with limited samples. Secondly， a self-supervised model based on multi-task meta-learning is constructed to realize online sorting of unlabeled signal streams. Finally， experimental results show that the proposed method achieves an accuracy of 95.82% on a simulated dataset at 4 dB， and it also performs well on real and public datasets， confirming the effectiveness of the proposed method in online sorting of data streams.

Key words: radar signal, online sorting, unsupervised learning, data stream

中图分类号:

TN974

普运伟,杜林,戴子瑜,何志强. 基于数据流自监督的雷达信号在线分选方法[J]. 吉林大学学报(工学版), 2026, 56(2): 564-574.

Yun-wei PU,Lin DU,Zi-yu DAI,Zhi-qiang HE. Online sorting method of radar signal based on self-supervised data stream[J]. Journal of Jilin University(Engineering and Technology Edition), 2026, 56(2): 564-574.

图/表 11

图1

图2

图3

图4

图5

表1

图6

表2

图7

表3

图8

参考文献 23

[1]	国强, 李明松, 周凯, 等. 基于势距图与改进云模型的多模雷达分选[J]. 吉林大学学报: 工学版, 2022, 52(8): 1904-1911.
	Guo Qiang, Li Ming-song, Zhou Kai, et al. Multi-mode radar sorting based on potential distance map and improved cloud model[J]. Journal of Jilin University (Engineering and Technology Edition), 2022, 52(8): 1904-1911.
[2]	Wang S Q, Gao C, Zhang Q, et al. Research progress on key technologies of radar signal sorting[J]. Recent Trends in Intelligent Computing, Communication and Devices, 2019, 1: 773-779.
[3]	Wang S Q, Hu G P, Zhang Q L, et al. The background and significance of radar signal sorting research in modern warfare[J]. Procedia Computer Science, 2019, 154: 519-523.
[4]	徐卓君, 杨雯婷, 杨承志, 等. 雷达脉内调制识别的改进残差神经网络算法[J]. 吉林大学学报: 工学版, 2021, 51(4): 1454-1460.
	Xu Zhuo-jun, Yang Wen-ting, Yang Cheng-zhi, et al. Improved residual neural network algorithm for radar intra-pulse modulation recognition[J]. Journal of Jilin University (Engineering and Technology Edition), 2021, 51(4): 1454-1460.
[5]	普运伟, 马蓝宇, 侯文太, 等. 模糊函数主脊切面特征提取的局域差分方法[J]. 数据采集与处理, 2019, 34(3): 386-395.
	Pu Yun-wei, Ma Lan-yu, Hou Wen-tai, et al. Loc-al differential method for feature extraction of main ridge section of ambiguity function[J]. Journal of Data Acquisition and Processing, 2019, 34(3): 386-395.
[6]	普运伟, 余永鹏, 姜萤, 等. 基于模糊函数多域特征融合与集成学习的雷达辐射源信号识别[J]. 控制与决策, 2024, 39(1):39-48.
	Pu Yun-wei, Yu Yong-peng, Jiang Ying, et al. Ra-dar emitter signal sorting based on multi-domain feature fusion of ambiguity function and ensemble learning[J]. Control and Decision, 2024, 39(1):39-48.
[7]	曹勇, 许秀颖, 蔡丹, 等. 玉米籽粒后熟品质性状主成分分析与综合评价[J]. 食品科学, 2024, 45(11): 1-7.
	Cao Yong, Xu Xiu-ying, Cai Dan, et al. Principal component analysis and comprehensive evaluation of post-ripening quality traits of corn kernels[J]. Food Science, 2024, 45(11): 1-7.
[8]	普运伟, 陈新杰, 余永鹏, 等. 基于模糊函数多维结构度量特征的雷达辐射源信号流在线分选[J]. 仪器仪表学报, 2023, 44(8):277-288.
	Pu Yun-wei, Chen Xin-jie, Yu Yong-peng, et al. Radar emitter signal online sorting based on multid-imensional structural metric features of ambiguity function[J]. Chinese Journal of Scientific Instrument, 2023, 44(8): 277-288.
[9]	Lang P, Fu X J, Cui Z D, et al. Subspace decomposition based adaptive density peak clustering for radar signals sorting[J]. IEEE Signal Processing Letters, 2022, 29: 424-428.
[10]	Jang H, Lee H, Shin J. Unsupervised meta-learning via fewshot pseudo-supervised contrastive learning[J]. Arxiv Preprint, 2023, 3: No.230300996.
[11]	Halidou A, Mohamadou Y, Ari A A A, et al. Review of wavelet denoising algorithms[J]. Multimed Tools Appl, 2023, 82: 41539-41569.
[12]	Goodfellow I J, Pouget A J, Mirza M, et al. Ge-nerative adversarial networks[C]∥Proceedings of the 27th International Conference on Neural Information Processing Systems, Montreal, Canada, 2014: 2672-2680.
[13]	王启帆. 基于数据增强的小样本学习方法研究[D]. 北京: 北京交通大学计算机学院, 2023.
	Wang Qi-fan. Research on few-shot learning method based on data augmentation[D]. Beijing: School of Computer Science, Beijing Jiaotong University, 2023.
[14]	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]∥ Proceedings of the 31st International Conference on Neural Information Processing Systems, New York, USA, 2017:6000-6010.
[15]	隋金坪. 基于数据流聚类的雷达在线分选方法研究[D]. 长沙:国防科技大学电子科学学院, 2020.
	Sui Jin-ping. Research on online sorting method of radar emitter based on data stream clustering[D]. Changsha: College of Electronic Science, National University of Defense Technology, 2020.
[16]	吴瑞琪, 周毅. 知识融入多源多任务学习的眼底图像分类方法[J]. 计算机工程与应用, 2025, 61(7): 255-266.
	Wu Rui-qi, Zhou Yi. Fundus image classification method with knowledge integration in multi-source multi-task learning[J]. Computer Engineering and Applications, 2025, 61(7): 255-266.
[17]	李红光, 王玉峰, 杨丽春. 基于元学习的小样本遥感图像目标检测[J]. 北京航空航天大学学报, 2024, 50(8):2503-2513.
	Li Hong-guang, Wang Yu-feng, Yang Li-chun. Few-shot remote sensing image object detection based on meta-learning[J]. Journal of Beijing University of Aeronautics and Astronautics, 2024, 50(8): 2503-2513.
[18]	Wang X, Fu X, Dong J, et al. Dynamic mo-dified chaotic particle swarm optimization for radar signal sorting[J]. IEEE Access, 2021, 9: 88452-88466.
[19]	Zhang D, Luo K. Clustering algorithm based on artificial bee colony optimization [C]∥Proceed-ings of the 2015 International Conference on Applied Science and Engineering Innovation, Jinan, China, 2015: 126-131.
[20]	吴婷. 基于EM算法的高斯混合模型在鸢尾花数据集的应用[J]. 网络安全技术与应用, 2022, 2022(4): 47-49.
	Wu Ting. Application of gaussian mixture model based on EM algorithm in Iris dataset[J]. Network Security Technology and Application, 2022, 2022(4): 47-49.
[21]	Feng S Y, Wang H X.Comparison of PCA and LDA dimensionality reduction algorithms based on wine dataset[C]∥33rd Chinese Control and Decision Conference (CCDC), Kunming, China, 2021: 1-6.
[22]	Zhang Z Z. Variance-SNR based noise suppression on linear canonical choi-williams distribute-on of LFM signals[J]. Chinese Journal of Electronics, 2022, 31(5): 804-820.
[23]	李思源, 徐天吉. 基于Wigner-Ville分布与Chrip-Z变换的高分辨时频分析方法[J]. 石油地球物理勘探, 2022, 57(1): 168-175.
	Li Si-yuan, Xu Tian-ji. High-resolution time-frequency analysis method based on Wigner-Ville distribution and Chrip-Z transform[J]. Geophysical Prospecting for Petroleum, 2022, 57(1): 168-175.

相关文章 14

[1]	康耀龙,冯丽露,张景安. 基于谱聚类欠取样的不平衡漂移大数据流分类算法[J]. 吉林大学学报(工学版), 2026, 56(2): 509-515.
[2]	王鹏,杨国栋. 面向嵌入式多核系统的缓存调度算法优化[J]. 吉林大学学报(工学版), 2024, 54(8): 2282-2287.
[3]	康耀龙,冯丽露,张景安,陈富. 基于谱聚类的高维类别属性数据流离群点挖掘算法[J]. 吉林大学学报(工学版), 2022, 52(6): 1422-1427.
[4]	魏晓辉,苗艳微,王兴旺. Rhombus sketch：自适应和准确的流数据sketch[J]. 吉林大学学报(工学版), 2022, 52(4): 874-884.
[5]	陈涛,高佳旭,张颖. 基于调制宽带转换器技术的侦察干扰一体化设计[J]. 吉林大学学报(工学版), 2022, 52(4): 942-949.
[6]	陈菲,杨峥,张志成,罗巍. 面向无标签数据的旋转机械故障诊断方法[J]. 吉林大学学报(工学版), 2022, 52(11): 2514-2522.
[7]	徐卓君,杨雯婷,杨承志,田彦涛,王晓军. 雷达脉内调制识别的改进残差神经网络算法[J]. 吉林大学学报(工学版), 2021, 51(4): 1454-1460.
[8]	张笑东,夏筱筠,吕海峰,公绪超,廉梦佳. 大数据网络并行计算环境中生理数据流动态负载均衡[J]. 吉林大学学报(工学版), 2020, 50(1): 247-254.
[9]	罗养霞, 郭晔. 基于数据依赖特征的软件识别[J]. 吉林大学学报(工学版), 2017, 47(6): 1894-1902.
[10]	杨静, 赵家石, 张健沛. 基于Johnson-Lindenstrauss转换的数据流降维方法[J]. 吉林大学学报(工学版), 2013, 43(06): 1626-1630.
[11]	随阳轶, 林君, 张晓拓,王世隆. 数据流可视化语言的运行模型[J]. 吉林大学学报(工学版), 2009, 39(06): 1635-1640.
[12]	王冠军, 马光胜, 李光顺. 高层次数据流低功耗综合[J]. 吉林大学学报(工学版), 2009, 39(05): 1331-1336.
[13]	随阳轶，林君，范永开 . 面向虚拟仪器的DFVPL：LabScene的自动连线[J]. 吉林大学学报(工学版), 2007, 37(05): 1203-1208.
[14]	张乐年，王凯. 杏仁在线分选系统的设计[J]. 吉林大学学报(工学版), 2005, 35(01): 97-0100.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

数据集		PSO-K	ABC-K	本文
Iris	Acc/%	88.24	86.32	93.48
Iris	T/s	6.24	4.43	2.67
Wine	Acc/%	72.12	61.89	94.65
Wine	T/s	7.99	4.45	2.79

模型	特征1和特征2		特征3和特征4
模型	Acc/%	T/s	Acc/%	T/S
SCMTL	29.12	3.78	34.43	2.36
GAN-SCMTL	60.43	7.86	69.78	4.54
BFFGAN-SCMTL	81.66	5.58	95.82	2.87

雷达信号	参数
雷达信号	RF/MHz	PW/μs
1	9 645、9 662、9 682、9 750、9 810 共5个频点波位组变，频率分集	20
2	9 762、9 773、9 792、9 807、9 822、 9 833共6个频点波位组变	3~5个脉冲为1组，每组PW在7、13任意
3	9 500~9 700单脉冲捷变	3~5个脉冲为1组，每组PW在0.9、1.0、 1.1、1.2任意
4	9 850固定	16
5	9 513/9 518/9 523/9 548/9 553/ 9 563共6个频点波位组变	3~5个脉冲为1组，每组PW在6、12、18任意