基于数据场联合决策图改进的GMM聚类

doi:10.12305/j.issn.1001-506X.2022.09.06

摘要/Abstract

摘要：

针对传统聚类方法在处理复杂电磁环境下的雷达信号时存在的聚类质量低、参数需要人为设置、易受孤立噪声脉冲干扰等问题, 提出一种基于数据场联合决策图改进的高斯混合模型(Gaussian mixture model, GMM)聚类算法。将数据场理论应用于数据对象密集程度的表征, 生成势能距离决策图, 进而自动实现聚类数目和中心点的选择, 最后结合GMM聚类实现对数据对象的聚类划分。仿真实验结果表明, 在脉冲到达角、脉宽、载频等参数存在较大抖动, 测量误差以及存在孤立噪声脉冲干扰和脉冲丢失时, 本文方法相较于现有典型分类方法具有更好的分选效果。

关键词: 雷达信号分选, 数据场, 决策图, 高斯混合模型聚类

Abstract:

Aiming at the problems of traditional clustering methods in processing radar signals in complex electromagnetic environment, such as low clustering quality, manual parameter setting and poor tolerance to isolated noise pulses, an improved Gaussian mixture model (GMM) clustering algorithm based on data field and decision graph is proposed. The data field theory is applied to the representation of the density of data objects, the potential energy distance decision graph is generated, and then the selection of cluster number and center point is realized automatically. Finally, the clustering division of data objects is realized combined with GMM. The simulation results show that this method has better sorting effect than the existing typical classification methods when there are significant jitter and measurement error in direction of arrival, pulse width and radio frequency, with the isolated noise pulses interference and pulse loss in the meantime.

Key words: radar signal classification, data field, decision diagram, Gaussian mixture model (GMM) clustering

中图分类号:

TN971

王磊, 张志勇, 曾维贵, 曹司磊, 张天赫. 基于数据场联合决策图改进的GMM聚类[J]. 系统工程与电子技术, 2022, 44(9): 2743-2751.

Lei WANG, Zhiyong ZHANG, Weigui ZENG, Silei CAO, Tianhe ZHANG. An improved GMM clustering based on data field and decision graph[J]. Systems Engineering and Electronics, 2022, 44(9): 2743-2751.

图/表 15

图1

图2

图3

图4

图5

图6

图7

表1

表2

图8

表3

图9

图10

图11

图12

参考文献 31

1	PINOU D, CHY R, HAYAJNEH T. Electromagnetic warfare and the cybersecurity threat[C]//Proc. of the IEEE 15th International Conference on Dependable, Autonomic and Secure Computing, 2017: 669-674.
2	WILLETTS B, RITCHIE M, GRIFFITHS H. Optimal time-frequency distribution selection for LPI radar pulse classification[C]//Proc. of the IEEE International Radar Conference, 2020: 327-332.
3	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. doi: 10.1049/joe.2019.0579
4	ZHANG C J , LIU Y C , SI W J . Synthetic algorithm for de-interleaving radar signals in a complex environment[J]. IET Radar, Sonar & Navigation, 2020, 14 (12): 1918- 1928.
5	JIANG W , FU X J . Improved algorithm for de-interleaving radar signals with overlapping features in the dynamically varying electromagnetic environment[J]. IET Radar, Sonar & Navigation, 2020, 14 (9): 1328- 1337.
6	MA Y H, LI Y, ZHU M T, et al. Radar working mode recognition based on hierarchical feature representation and clustering[C]//Proc. of the IET International Radar Conference, 2021: 1629-1633.
7	淦文燕, 李德毅, 王建民. 一种基于数据场的层次聚类方法[J]. 电子学报, 2006, 34 (2): 258- 262.
	GAN W Y , LI D Y , WANG J M . An hierarchical clustering method based on data fields[J]. Acta Electronic Sinica, 2006, 34 (2): 258- 262.
8	ZHANG H, WEI H S, SHEN Y C, et al. Bottlenecks and feasible solutions of data field clustering in impact factor, time resolution, selecting core objects and merging process[C]//Proc. of the International Conference on Artificial Intelligence and Security, 2019: 106-120.
9	CHEN J, ZHANG J Y, WU J H, et al. Review on the research of K-means clustering algorithm in big data[C]//Proc. of the IEEE 3rd International Conference on Electronics and Communication Engineering, 2020: 107-111.
10	SRIDEVI K N, PRAKASHA S. Comparative study on various clustering algorithms review[C]//Proc. of the IEEE 5th International Conference on Intelligent Computing and Control Systems, 2021: 153-158.
11	赵贵喜, 刘永波, 王岩, 等. 数据场和K-means算法融合的雷达信号分选[J]. 雷达科学与技术, 2016, 14 (5): 517- 525. doi: 10.3969/j.issn.1672-2337.2016.05.012
	ZHAO G X , LIU Y B , WANG Y , et al. Radar signal sorting based on data field and K-means clustering fusion algorithm[J]. Radar Science and Technology, 2016, 14 (5): 517- 525. doi: 10.3969/j.issn.1672-2337.2016.05.012
12	张怡霄, 郭文普, 康凯, 等. 基于数据场联合PRI变换与聚类的雷达信号分选[J]. 系统工程与电子技术, 2019, 41 (7): 1509- 1515.
	ZHANG Y X , GUO W P , KANG K , et al. Radar signal sorting method based on data field combined PRI transform and clustering[J]. Systems Engineering and Electronics, 2019, 41 (7): 1509- 1515.
13	YOUSEFI T , ODABAS M S , OKTAS R . Overview of different methods used in clustering algorithms[J]. Black Sea Journal of Engineering and Science, 2020, 3 (4): 173- 189.
14	沙作金. 数据场聚类与平面变换雷达信号分选算法研究[D]. 哈尔滨: 哈尔滨工程大学, 2019.
	SHA Z J. Research on data field clustering and planar transform radar signal sorting algorithm[D]. Harbin: Harbin Engineering University, 2019.
15	郜丽鹏, 沙作金. 一种改进的数据场聚类算法[J]. 应用科技, 2019, 46 (6): 30- 34.
	GAO L P , SHA Z J . An improved data field clustering algorithm[J]. Applied Science and Technology, 2019, 46 (6): 30- 34.
16	RODRIGUEZ A , LAIO A . Clustering by fast search and find of density peaks[J]. Science, 2014, 344 (6191): 1492- 1496. doi: 10.1126/science.1242072
17	ZHU X H , SHANG J L , SUN Y , et al. PSO-CFDP: a particle swarm optimization-based automatic density peaks clustering method for cancer subtyping[J]. Human Heredity, 2019, 84 (1): 9- 20. doi: 10.1159/000501481
18	LI H B, ZHAO J, ZHANG Y. Signals deinterleaving for ES systems using improved CFSFDP algorithm[C]//Proc. of the IEEE Radar Conference, 2019: 1-5.
19	CAROTENUTO V , MAIO A D . A clustering approach for jamming environment classification[J]. IEEE Trans.on Aerospace and Electronic Systems, 2021, 57 (3): 1903- 1918. doi: 10.1109/TAES.2021.3050655
20	SUI J P , LIU Z , LIU L , et al. Online non-cooperative radar emitter classification from evolving and imbalanced pulse streams[J]. IEEE Sensors Journal, 2020, 20 (14): 7721- 7730. doi: 10.1109/JSEN.2020.2981976
21	胡建华, 尹慧琳. 基于PSO-TVAC的中心自适应权的FCM聚类算法[J]. 应用数学进展, 2021, 10 (4): 953- 962.
	HU J H , YIN H L . FCM clustering algorithm based on PSO-TVAC algorithm with adaptively weighted centers[J]. Advances in Applied Mathematics, 2021, 10 (4): 953- 962.
22	DENG D S. DBSCAN clustering algorithm based on density[C]//Proc. of the IEEE 7th International Forum on Electrical Engineering and Automation, 2020: 949-953.
23	WANG C X , JI M , WANG J , et al. An improved DBSCAN method for LiDAR data segmentation with automatic eps estimation[J]. Sensors, 2019, 19 (1): 172- 198. doi: 10.3390/s19010172
24	BELLAVISTA P, CAMPESTRI M, FOSCHINI L, et al. Clustering of spatial data with DBSCAN: an assessment of stark[C]//Proc. of the IEEE Symposium on Computers and Communications, 2019: 1-6.
25	MA Y H , HAO Y . Antenna classification using gaussian mixture models (GMM) and machine learning[J]. IEEE Open Journal of Antennas and Propagation, 2020, 1 (1): 320- 328.
26	GONG X H, MENG H D, WANG X Q. A GMM-based algorithm for classification of radar emitters[C]//Proc. of the IEEE 9th International Conference on Signal Processing, 2008: 2543-2546.
27	MURAT S , NECMETTIN Y , BUNYAMIN Y . An improvement on Fibonacci search method in optimization theory[J]. Applied Mathematics and Computation, 2004, 147 (3): 893- 901. doi: 10.1016/S0096-3003(02)00828-7
28	PATRANABIS D , PAUL A , DEY A . Efficient enhancement/lowering of integrator and differentiator time constants with Fibonacci numbers helping their evaluation[J]. IEEE Trans.on Circuits & Systems, 1982, 29 (6): 409- 411.
29	FISHER P S , KOHLBECKER E E . A generalized Fibonacci sequence[J]. The American Mathematical Monthly, 1961, 68 (5): 455- 459. doi: 10.1080/00029890.1961.11989696
30	HOU X D , ZHANG T , XIONG G , et al. A novel steganalysis framework of heterogeneous images based on GMM clustering[J]. Signal Processing: Image Communication, 2014, 29 (3): 385- 399. doi: 10.1016/j.image.2014.01.006
31	BALA R , SIKKA S , SINGH J . A comparative analysis of clustering algorithms[J]. International Journal of Computer Applications, 2014, 100 (15): 35- 39. doi: 10.5120/17603-8293

目标	到达角参数/(°)		载频参数/MHz		脉宽参数/μs		脉冲数
目标	中心值	方差	中心值	方差	中心值	方差	脉冲数
1	15	3.2	9 398.7	4.0	0.1	0.04	396
2	15	3.8	9 390.5	0.6	1.5	0.04	134
3	27	4.3	9 392.3	0.8	1.1	0.02	156
4	31	3.7	9 389.0	0.7	1.2	0.03	136
5	32	1.3	9 417.2	4.0	0.1	0.04	339
6	40	1.9	9 415.9	2.9	0.2	0.01	207
7	43	1.2	9 374.9	2.2	0.3	0.01	136
8	45	2.0	9 368.3	2.2	0.3	0.03	160
9	46	3.9	9 371.8	1.2	0.7	0.03	160
10	47	3.3	9 419.3	1.5	0.5	0.03	159
11	48	2.5	9 418.6	1.0	0.9	0.04	108
12	50	1.3	9 414.7	0.7	1.3	0.05	62
13	51	3.5	9 387.7	0.5	0.1	0.04	396

环境/参数		参考值
硬件平台	CPU	Intel(R) Core(TM) i5-3317U
	主频	1.7 GHz
	内存	10 GHz 2 400 MHz
软件平台	系统	WIN10 64位专业版
软件平台	软件	Matlab 2020a
算法参数	辐射因子σ	0.01
算法参数	截断距离d_c	0.01

聚类算法	正确聚类平均数	聚类正确率/%
改进K-means	2 519.17	98.83
改进数据场聚类	2 515.61	98.69
改进CFSFDP	2 523.76	99.01
本文方法	2 531.92	99.33

[1]	张怡霄, 郭文普, 康凯, 姚云龙, 张林科, 张薇. 基于数据场联合PRI变换与聚类的雷达信号分选[J]. 系统工程与电子技术, 2019, 41(7): 1509-1515.
[2]	凌牧, 袁海文, 马钊, 黄操. 改进的动态故障树转化为二元决策图的成分组合算法与应用[J]. 系统工程与电子技术, 2016, 38(7): 1600-1605.
[3]	陈瑶，李峭，赵长啸，熊颖. 基于OBDD的航空电子网络可靠性分析[J]. Journal of Systems Engineering and Electronics, 2013, 35(1): 230-236.
[4]	赵勃，肖宇峰，刘岩. 基于OBDD的通信网链路重要性评估[J]. Journal of Systems Engineering and Electronics, 2011, 33(10): 2348-2352.
[5]	刘旭波，司锡才. 雷达信号分选实现的新方法[J]. Journal of Systems Engineering and Electronics, 2010, 32(1): 53-56.