交叉验证的GRNN神经网络雷达目标识别方法研究

doi:10.3969/j.issn.1009-086x.2018.04.018

现代防御技术 ›› 2018, Vol. 46 ›› Issue (4): 113-119.DOI: 10.3969/j.issn.1009-086x.2018.04.018

交叉验证的GRNN神经网络雷达目标识别方法研究

林悦¹, 夏厚培²

1.南京信息工程大学电子与信息工程学院,江苏南京 210044;
2.中国船舶重工集团公司第七二四研究所,江苏南京 211106

收稿日期:2017-10-18 修回日期:2017-12-13 出版日期:2018-08-30 发布日期:2020-11-25
作者简介:林悦(1993-),女,江苏南京人。硕士生,主要研究方向为雷达目标识别技术。通信地址:211106 江苏省南京市江宁区水阁路长青街30号 E-mail:lin-yue-93@163.com

Radar Target Recognition Based on Cross Validation GRNN

LIN Yue¹, XIA Hou-pei²

1. Nanjing University of Information Science & Technology,College of Electrics and Information Engineering,Jiangsu Nanjing 210044,China;
2. China Shipbuilding Industry Corporation,724 Research Institute,Jiangsu Nanjing 211106,China

Received:2017-10-18 Revised:2017-12-13 Online:2018-08-30 Published:2020-11-25

摘要/Abstract

摘要： 雷达目标识别是现代雷达技术的重要发展方向之一,在未来武器系统中具有重要的意义。针对高分辨一维距离像在预处理中平移敏感性的问题,使用了全局最小熵距离对齐算法,能够较准确快速对准距离像单元,提高了距离对齐的精度。为了提高雷达目标识别的准确率,提出了一种与广义回归神经网络模型(generalized regression neural network,GRNN)相结合的目标识别方法。利用K重交叉验证法对神经网络训练,并且根据最小均方误差寻找出GRNN神经网络光滑因子spread的最优值,同时获得目标识别训练样本的最优输入输出值。通过对比,取得最优光滑因子的GRNN神经网络将大幅度提高其收敛速度与泛化能力。仿真实验证明,基于改进GRNN神经网络的雷达目标识别可以获得较高较稳定的识别正确率。

关键词: 雷达自动目标识别, 高分辨距离像, 最小熵对齐法, 广义回归神经网络, 光滑因子, 交叉验证法

Abstract: As one of the important development directions of modern radar technology, radar target recognition is of great significance to the future weapon system. Aiming at the problem of translational sensitivity in high resolution range profile (HRRP) in preprocessing, the global minimum entropy distance alignment algorithm is used to accurately align the range profile and improve the accuracy of distance alignment. To improve the accuracy of radar target recognition, a new method of target recognition based on generalized regression neural network (GRNN) is proposed. The K-fold cross-validation method is used to train the neural network, and the optimal value of the smoothing factor spread of GRNN neural network is found according to the mean square error (MSE). The optimal input and output values of the target recognition training samples are obtained. By contrast, GRNN neural network with optimal smoothing factor spread will greatly improve its convergence speed and generalization ability. The simulation results show that the radar target recognition based on the improved GRNN neural network can obtain higher and more accurate recognition accuracy.

Key words: radar automatic target recognition(RATR), high resolution range profile(HRRP), minimum entropy alignment algorithm, generalized regression neural network(GRNN), spread, cross validation

中图分类号:

TN959
TP183

林悦, 夏厚培. 交叉验证的GRNN神经网络雷达目标识别方法研究[J]. 现代防御技术, 2018, 46(4): 113-119.

LIN Yue, XIA Hou-pei. Radar Target Recognition Based on Cross Validation GRNN[J]. Modern Defense Technology, 2018, 46(4): 113-119.

参考文献

[1] 刘宏伟,杜兰,袁莉,等.雷达高分辨距离像目标识别研究进展[J].电子与信息学报,2005,27(8):1328-1334.
LIU Hong-wei,DU Lan,YUAN Li,et al.Progress in Radar Automatic Target Recognition Based on High Range Resolution Profile [J].Journal of Electronics & Information Technology,2005,27(8):1328-1334.
[2] DU Lan,LIU Hong-wei,BAO Zheng.Radar HRRP Statistical Recognition Based on Hypersphere Model [J].Signal Processing,2007,88(5):1176-1190.
[3] 丛瑜.基于核方法的高分辨雷达目标特征提取与识别[D].长沙:国防科学技术大学,2006.
CONG Yu.The Target Feature Extraction and Recognition for High Range Recognition Radar Based on the Kernel Method[D].Changsha:National University of Defense Technology,2006.
[4] BOTHA E C,BARNARD E,BARNARD C J.Feature Based Classification of Aerospace Radar Target Using Neural Networks[J].Neural Networks,1996,9 (1):129-142.
[5] 林喆,兰生,张宇航.基于广义回归神经网络的油纸绝缘变压器的寿命预测[J].高压电器,2015,51(2):125-130.
LIN Zhe,LAN Sheng,ZHANG Yu-hang.Life Prediction of Oil Paper Insulation of Transformer Based on GRNN Neural Network[J].High Voltage Apparatus,2015,51(2):125-130.
[6] MASON S J,TIPPETT M K.Cross-Validation for Selection of Statistical models [C]//17th Conference on Probability and Statistics in the Atmospheric Sciences,Seattle,USA,January 10-15,2004.
[7] ZHU Dai-yin,WANG Ling,TAO Qing-nian,et al.ISAR Range Alignment by Minimizing the Entropy of the Average Range Profile [C]//2006 IEEE Conference on Radar,Verona,USA,April 24-27,2006.
[8] ZHU Dai-yin,WANG Ling,YU Yu-sheng,et al.Robust ISAR Range Alignment via Minimizing the Entropy of the Average Range Profile [J].IEEE Geoscience and Remote Sensing Letters,2009,6(2):204-208.
[9] SPECHT D F.A General Regression Neural Network[J].IEEE Transactions on Neural Network,1991,2(6):568-576.
[10] 范国清,王威,郗晓宁.基于广义回归神经网络的电离层VTEC建模[J].测绘学报,2010,39(1):16-21.
FAN Guo-qing,WANG Wei,XI Xiao-ning.Modeling of Ionosphere VTEC Using Generalized Regression Neural Network [J].Acta Geodaetica et Cartographica Sinica,2010,39(1):16-21.
[11] 周敏,李世玲.广义回归神经网络在非线性系统建模中的应用[J].计算机测量与控制,2007,15(9):1189-1191.
ZHOU Min,LI Shi-ling.Application of GRNN and Uniform Design to Nonlinear System Modeling [J]. Computer Measurement & Control,2007,15(9):1189-1191.
[12] CHENG Jia-xing,ZHANG Qi,WU Xu,et al.An Application of Minimum Penalty Coefficient K-Fold Cross Validation-Support Vector Machine in the Regression Analysis of Railway Monthly Freight Volume[C]//International Conference on Advanced Computer Science and Engineering (ACSE),Guangzhou,China,June 29,2014.
[13] LEUNG H,HENNESSEY G,DROSOPOULOS A.Signal Detection Using the Radial Basis Function Coupled Map Lattice[J].IEEE Transactions on Neural Networks,2000,11 (5):1133-1151.
[14] 柴晶.雷达高分辨距离像目标识别的拒判算法和特征提取技术研究[D].西安:西安电子科技大学,2010.
CHAI Jing.Study of Rejection Algorithm and Feature Extraction Technique on Radar HRRP Target Recognition [D].Xi'an:Xidian University,2010.
[15] 廖阔.基于高分辨距离像的雷达自动目标识别技术研究[D].成都:电子科技大学,2012.
LIAO Kuo.Study of Radar Automatic Target Recognition Base on High Range Resolution Profile[D].Chengdu:University of Electronic Science and Technology of China,2012.

交叉验证的GRNN神经网络雷达目标识别方法研究

Radar Target Recognition Based on Cross Validation GRNN

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	闫善勇. 基于灰色理论的雷达目标与IFF点迹关联方法[J]. 现代防御技术, 2023, 51(4): 46-52.
[2]	肖志河, 任红梅, 董纯柱, 赵涛, 陈轩. 空天目标电磁散射特性建模与特征认知研究[J]. 现代防御技术, 2023, 51(3): 83-90.
[3]	孙涛, 龙庆华, 巢育龙, 董博, 罗军. 防空反导雷达高作战效能人机系统设计方法[J]. 现代防御技术, 2022, 50(4): 107-115.
[4]	翟心蝶, 杨刚, 廉杰. 基于二阶Keystone的微弱运动目标检测[J]. 现代防御技术, 2021, 49(3): 105-114.
[5]	王志虎, 沈小青, 桂伟龙. 光学成像小目标检测技术综述[J]. 现代防御技术, 2020, 48(5): 67-73.
[6]	陈睿容, 孙武, 贾学振, 徐秋锋. 基于二阶Keystone的空间目标检测方法[J]. 现代防御技术, 2020, 48(4): 93-101.
[7]	王胜喜, 曾小东. 机载平台电磁兼容性预测与设计[J]. 现代防御技术, 2020, 48(2): 89-95.
[8]	沈小玲, 冯孝斌, 张达凯, 刘胤凯, 张云, 冯雨. 提高近场测量系统隔离度的方法研究[J]. 现代防御技术, 2020, 48(1): 95-99.
[9]	罗木生, 侯学隆, 王宗杰, 王培源. 反舰导弹对岛礁区目标的搜捕概率研究[J]. 现代防御技术, 2018, 46(6): 30-35.
[10]	林菡, 李昌玺, 陈丽娟. MIMO-FNN模型的弹道导弹目标识别方法[J]. 现代防御技术, 2018, 46(6): 36-43.
[11]	杨军佳, 叶晨亮, 刘志青, 廉博. SPA和DS证据理论的多传感器航迹关联方法[J]. 现代防御技术, 2018, 46(6): 75-79.
[12]	唐玉文, 何明浩, 韩俊, 张小涵. 机载火控雷达典型空-空工作模式浅析[J]. 现代防御技术, 2018, 46(6): 87-93.
[13]	董爱, 夏芒, 李陟. 基于距离分辨的多目标同一性识别方法[J]. 现代防御技术, 2018, 46(3): 98-104.
[14]	单时卓, 张南, 兰洪光, 张艳. 导弹武器系统精度类数据误差统计方法研究[J]. 现代防御技术, 2018, 46(3): 133-138.
[15]	王万田, 袁俊泉, 王力宝, 陈阿磊. 基于MDCFT的天空双基地雷达机动目标检测方法[J]. 现代防御技术, 2018, 46(3): 191-198.