面向非协作多功能雷达的波形单元提取方法

doi:10.12305/j.issn.1001-506X.2021.10.18

Abstract

Abstract:

Aiming at the problems that the current waveform unit extraction technology is difficult to apply in the reconnaissance data of non-cooperative multi-function radar (MFR), a binary classification model based on MFR multi-parameter sequence is constructed, and a density clustering algorithm with adaptive input parameters for classification is proposed. This method does not need to rely on the prior knowledge of MFR waveform library, and uses unsupervised learning to extract waveform units. At the same time, making full use of the joint changes among multiple parameters and the overall distribution information of the data set to improve the robustness of the algorithm. The performance of the algorithm can be adjusted for different user requirements by setting the input parameter λ. Simulation results show that the proposed algorithm can effectively extract non-cooperative MFR waveform units, and can adapt to the interference caused by measurement error and pulse loss. It has good robustness and accuracy, and is conductive to practical engineering application.

Key words: non-cooperative multi-function radar, waveform unit, data-driven, unsupervised learning, density-based spatial clustering

CLC Number:

TN95

Liu YANG, Weigang ZHU, Shouye LYU, Shuang MA. Waveform unit extraction method for non-cooperative multi-function radar[J]. Systems Engineering and Electronics, 2021, 43(10): 2843-2850.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Table 1

Table 2

Table 3

Fig.6

Fig.7

Fig.8

Fig.9

Table 4

References 15

1	HAYSTEAD J . DARPA seeks proposals for adaptive radar countermeasures[J]. Journal of Electronic Defense, 2012, 35.
2	张双图. 雷达功能推断方法研究及软件设计[D]. 西安: 西安电子科技大学, 2017.
	ZHANG S T. Study on the methods of radar function inference and software design[D]. Xi'an: Xidian University, 2017.
3	VISNEVSKI N, HAYKIN S, KRISHNAMURTHY V, et al. Hidden Markov models for radar pulse train analysis in electronic warfare[C]//Proc. of the IEEE International Conference on Acoustics, Speech & Signal Processing, 2005: 597-600.
4	刘海军. 雷达辐射源识别关键技术研究[D]. 长沙: 国防科学技术大学, 2010.
	LIU H J. Researches on identification key technology for radar emitter[D]. Changsha: National University of Defense Technology, 2010.
5	陈维高, 朱卫纲, 唐晓婧, 等. 栈式降噪自编码器在波形单元识别中的应用[J]. 哈尔滨工业大学学报, 2018, 50 (11): 94- 100.
	CHEN W G , ZHU W G , TANG X J , et al. The application ofstacked denoising autoencoders in waveform unit identification[J]. Journal of Harbin Institute of Technology, 2018, 50 (11): 94- 100.
6	JEROME D V. Tracking the operational mode of multi-fuction radars[D]. Canada: Mcmaster, 2007.
7	马爽. 多功能雷达电子情报信号处理关键技术研究[D]. 长沙: 国防科学技术大学, 2013.
	MA S. Research on ELINT signal processing key technologies for multifunction radar[D]. Changsha: National University of Defense Technology, 2013.
8	方旖, 陈秋菊, 潘继飞, 等. 基于贝叶斯的多功能雷达脉冲列变化点检测[J]. 指挥与控制学报, 2019, 5 (4): 308- 315. doi: 10.3969/j.issn.2096-0204.2019.04.0308
	FANG Y , CHEN Q J , PAN J F , et al. Multi-function radar pulse sequence change point detection based on bayesian criterion[J]. Journal of Command and Control, 2019, 5 (4): 308- 315. doi: 10.3969/j.issn.2096-0204.2019.04.0308
9	王勇军. 一种改进的事件驱动的MFR雷达字提取方法[J]. 现代雷达, 2019, 41 (3): 17- 20.
	WANG Y J . Novel approach of radar word extraction for MFRs based on event-driven method[J]. Modern Radar, 2019, 41 (3): 17- 20.
10	COCHRAN D , SUVOROVA S , HOWARD S D , et al. Waveform libraries[J]. IEEE Signal Processing Magazine, 2009, 26 (1): 12- 21. doi: 10.1109/MSP.2008.930415
11	KAUPPI J , MARTIKAINEN K , RUOTSALAINEN U . Hierarchical classification of dynamically varying radar pulse repetition interval modulation patterns[J]. Neural Networks, 2010, 23 (10): 1226- 1237. doi: 10.1016/j.neunet.2010.06.008
12	王实美. 基于DBSCAN的自适应非均匀密度聚类算法研究[D]. 北京: 北京交通大学, 2017.
	WANG S M. Research on adaptive varied density clustering algorithm based on DBSCAN[D]. Beijing: Beijing Jiaotong University, 2017.
13	李文杰, 闫世强, 蒋莹, 等. 自适应确定DBSCAN算法参数的算法研究[J]. 计算机工程与应用, 2019, 55 (5): 1- 7, 148.
	LI W J , YAN S Q , JIANG Y , et al. Research on method of self-adaptive determination of DBSCAN algorithm parameters[J]. Computer Engineering and Applications, 2019, 55 (5): 1- 7, 148.
14	周志华. 机器学习[M]. 北京: 清华大学出版社, 2016.
	ZHOU Z H . Machine learning[M]. Beijing: Tsinghua University Press, 2016.
15	STEINBACH M, KARYPIS G, KUMAR V. A comparison of document clustering techniques[C]//Proc. of the Knowledge Discovery and Data Mining Workshop on Text Mining, 2000: 525-526.

类型	波形单元	脉冲重复间隔PRI/μs		载频RF/MHz		脉宽PW/μs		脉冲个数
类型	波形单元	类型	取值	类型	取值	类型	取值	脉冲个数
A	1	参差	(880/850/940)×3	固定	3 423	固定	15.1	9
	2		(846/783/723)×3		3 365		10.0	9
	3		(1 060/1 150/1 100)×3		3 236		29.4	9
	4		(1 200/1 250/1 310)×3		3 248		55.1	9
B	5~34	滑变	490~620	捷变	3 100~3 450	可选择	6.0/10.0	5~40
C	35	组变	745×4	固定	3 380	固定	6.0	4
	36		825×4		3 300		10.0	4
	37		1 276×4		3 221		15.1	4
	38		1 342×4		3 185		55.1	4
D	39	固定	955	跳变	3 400×12/3 360×12/3 320×12	可选择	10.0×12/15.1×12/29.4×12	36
E	40	抖动	1 150~1 450	固定	3 100	可选择	6.0/10.0	40

算法	是否采用二分类模型	Eps	MinPts	F-score
DA-DBSCAN	否	7.9	3	0.720
DA-DBSCAN	是	6.8	3	0.979
DBSCAN	否	6.0	6	0.642
		7.0	6	0.667
		8.0	6	0.731
	是	4.0	6	0.866
		5.5	6	0.943
		7.0	6	0.965
层次聚类	否	—	—	0.664
层次聚类	是	—	—	0.928

(RDP/EDL)/%	TPR/%	FPR/%	F-score
0	92.75/92.98	1.03/1.04	0.986 5/0.986 5
5	92.72/87.15	1.16/2.22	0.985 1/0.973 1
10	92.67/88.60	1.26/2.43	0.984 0/0.972 4
15	92.62/87.05	1.36/2.73	0.982 8/0.969 1
20	92.59/86.35	1.45/3.04	0.981 7/0.966 3
25	92.59/85.75	1.57/3.49	0.980 3/0.962 5
30	92.47/86.49	1.62/3.90	0.979 5/0.959 9
35	92.36/86.51	1.74/4.25	0.977 9/0.957 4
40	92.37/85.27	1.90/4.65	0.976 1/0.953 7
45	92.37/86.51	2.11/5.14	0.973 8/0.950 8
50	91.94/86.18	2.16/5.47	0.972 2/0.948 4

λ	TPR/%	FPR/%	F-score
-1	94.49	12.89	0.899 9
-0.9	94.20	11.89	0.906 4
-0.8	93.87	10.89	0.912 9
-0.7	93.50	9.90	0.919 4
-0.6	93.08	8.93	0.925 8
-0.5	92.66	7.94	0.932 5
-0.4	92.13	6.97	0.939 0
-0.3	91.54	6.05	0.945 3
-0.2	90.80	5.11	0.951 6
-0.1	89.87	4.21	0.957 6
0	88.59	3.36	0.963 3
0.1	86.70	2.54	0.968 3
0.2	83.50	1.79	0.971 9
0.3	78.87	1.18	0.973 3
0.4	72.02	0.71	0.971 6
0.5	63.34	0.39	0.966 9
0.6	53.11	0.00	0.959 2
0.7	41.34	0.00	0.948 2
0.8	27.71	0.00	0.933 0
0.9	12.99	0.00	0.913 4
1	0.11	0.00	—

[1]	Pengyu CAO, Chengzhi YANG, Limeng SHI, Hongchao WU. Unknown radar signal processing based on PSO-DBSCAN and SCGAN [J]. Systems Engineering and Electronics, 2022, 44(4): 1158-1165.
[2]	Jia HUANG, Sijiang CHANG. Data-driven method based impact time and impact angle control guidance law [J]. Systems Engineering and Electronics, 2022, 44(10): 3213-3220.
[3]	Chen MENG, Huahui YANG, Cheng WANG, Zheng MA. Review on data-driven fault diagnosis for electronic components and units level of weapon system [J]. Systems Engineering and Electronics, 2021, 43(2): 574-583.
[4]	Lei YANG, Xinyao MAO, Xiaowei YANG, Hai ZHANG, Fei YANG, Lin SUN. Dynamic evaluation of radar performance index based on unsupervised Bayesian learning [J]. Systems Engineering and Electronics, 2021, 43(1): 74-82.
[5]	Yixiao ZHANG, Wenpu GUO, Kai KANG, Yunlong YAO, Pan WANG. Key radar signal fast recognition method based on clustering and time-series correlation [J]. Systems Engineering and Electronics, 2020, 42(3): 597-602.
[6]	XU Jianguo, LI Mengjun, JIANG Jiang, YOU Hanlin. Data-driven motif analysis of technology breakthrough network [J]. Systems Engineering and Electronics, 2017, 39(5): 1072-1077.

Waveform unit extraction method for non-cooperative multi-function radar

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 13

References 15

Related Articles 6

Recommended Articles

Metrics

Comments