[1]方旖,毕大平,潘继飞,等.基于主成分分析的雷达行为状态聚类分析方法[J].探测与控制学报,2020,42(02):112.[doi:.]
 FANG Yi,BI Daping,PAN Jifei,et al.Radar Behavior State Cluster Analysis Based on Principal Component Analysis[J].,2020,42(02):112.[doi:.]
点击复制

基于主成分分析的雷达行为状态聚类分析方法()
分享到:

《探测与控制学报》[ISSN:1008-1194/CN:61-1316/TJ]

卷:
42
期数:
2020年02
页码:
112
栏目:
出版日期:
2020-04-15

文章信息/Info

Title:
Radar Behavior State Cluster Analysis Based on Principal Component Analysis
文章编号:
1008-1194(2020)02-0112-07
作者:
方旖1毕大平1潘继飞12陈秋菊12
1.国防科技大学电子对抗学院预警对抗系,安徽 合肥 230037;2.电子对抗信息处理实验室,安徽 合肥 230037
Author(s):
FANG Yi1 BI Daping1 PAN Jifei12 CHEN Qiuju12
1.Institute of Electronic Countermeasure, National University of Defence Technology, Hefei 230037, China; 2.Laboratory of Electronic Countermeasure Information Processing, National University of Defence Technology, Hefei 230037, China
关键词:
认知对抗雷达雷达行为C-均值算法主成分分析聚类识别
Keywords:
cognitive countermeasure radar radar behavior c-mean value algorithm principal component analysis clustering recognition
分类号:
TP301.6; TN957.51
DOI:
.
文献标志码:
A
摘要:
针对在实际作战中雷达对抗侦察设备对未知雷达进行工作状态识别时,难以获取和利用充足的训练样本的问题,提出了基于主成分分析的雷达行为状态聚类分析方法。该方法通过分析雷达行为状态类型与雷达信号脉冲参数间的本质关联,以获得可用于分类的特征。首先对机载侧视雷达远区搜索、近区搜索及跟踪等不同工作状态下的雷达信号特征进行深入分析,选择合适的雷达脉冲序列组成雷达样本库。然后,对测试信号特征进行预处理,结合主成分分析方法提取合适的特征参数,并对C-均值聚类算法进行了改进,完成了样本数据的分类,该方法能够在小样本数据条件下进行分类。仿真结果表明,该方法可以用于雷达行为状态的分类并且不易受初始数据及随机聚类中心的影响。
Abstract:
Aiming at the problem that it was difficult to acquire and utilize sufficient training samples when radar countermeasure reconnaissance equipment was used to identify the working state of unknown radar in actual combat, a method of radar behavior state cluster analysis based on principal component analysis was proposed. By analyzing the essential correlation between radar behavior state types and radar signal pulse parameters, this method obtained the characteristics that could be used for classification. Firstly, the signal characteristics of an airborne side-looking radar under different working conditions, such as far region search, near region search and tracking, were analyzed in depth. Then, the characteristics of the test signal were preprocessed, and the appropriate characteristic parameters were extracted by combining the principal component analysis method, and the c-mean clustering algorithm was improved to complete the classification of sample data. Simulation results showed that this method could be used to classify radar behavior and was not easily affected by initial data and random clustering center.

参考文献/References:

[1]戴幻尧,周波,雷昊.认知电子战的关键技术发展动态与分析[J].飞航导弹,2014(9):69-72.
[2]刘海军,樊昀,李悦等.多功能雷达建模中的雷达字提取技术研究[J].国防科技大学学报, 2010, 32(2): 91-96.
[3]Visnevski N, Haykin S, Krishnamurthy V, Dilkes A, et al. Hidden markov models for radar pulse train analysis in electronic warfare[C]//Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing. US: IEEE, 2005:597-600.
[4]Hassan H E, Chan F, Chan Y T. Joint deinterleaving / recognition of radar pulses [C]//IEEE Canadian Conference on Electrical and Computer Engineering. US: IEEE, 2003: 2009-2014.
[5]Li C, Wang W, Wang X S. A method for extracting radar words of multi-function radar at eata level[C]//IET International Radar Conference. UK: IET, 2013.
[6]李程. 雷达电子侦察的多层次处理关键技术研究[D]. 长沙: 国防科技大学, 2015.
[7]贾朝文,周水楼.机载雷达工作模式识别[J].电子信息对抗技术,2011(1):16-18+78.
[8]Ma S, Liu Z, Jiang W L. Reconstruction of multifunction radar search plan based on multiple sequence alignment[J]. Chinese Journal of Electronics, 2013,23(1): 213-218.
[9]李辉,郑坤,金炜东,等.基于多层次建模的机载相控阵雷达工作模式识别[J].电子信息对抗技术,2016,31(4):1-5+42.
[10]孟祥豪, 罗景青, 王杰贵. 基于自提取脉冲样本图的雷达信号快速提取法[J].航天电子对抗, 2014, 30(4): 53-57.
[11]Yang Z T, Qiu W, Sun H J. Robust radar emitter recognition based on the three-dimensional distribution feature and transfer learning[J]. Sensors, 2016,16: 289.
[12]代鹂鹏,王布宏,曹帅,等.基于最优解析树提取的多功能雷达状态快速估计方法[J].电子学报,2016,44(3):514-519.
[13]Thon M, Jaeger H. Links between multiplicity automata, observable operator models and predictive state representations: a unified learning framework [J]. Journal of Machine Learning Research, 2015(16): 103-147.
[14]Yun Taek Im,Seong Ook Park. An FMCW and chirp pulse-doppler radar system for surveillance in X-band[J]. Microwave and Optical Technology Letters, 2013, 55(3):474-479.
[15]王洪授,黄同愿,陈红光,等. 基于聚类主成分分析法的变压器参数关联度评估[J].火力与指挥控制,2017,42(4):133-137.
[16]陈志飞, 时宏伟, 吕学斌, 等. 基于均值漂移和模糊C均值聚类的图像分割算法[J]. 计算机应用与软件, 2013(11):13-17.

备注/Memo

备注/Memo:
收稿日期:2019-12-02
基金项目:国家自然科学基金项目资助(61671453)
作者简介:方旖(1995—),女,浙江杭州人,硕士研究生,研究方向:认知雷达对抗。E-mail: fangyi950129@163.com
更新日期/Last Update: 2020-05-15