采用最优集成学习的小样本电磁脉冲信号分类

摘要
图/表
参考文献(43)
相关文章 (7)

全文: PDF (840 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要闪电与核爆电磁脉冲分类是核检测系统中的关键问题，其主要难点便是其正负样本不均衡程度可达到，因此我们提出了一种名为SMALLBAG的集成学习分类方法。针对小样本问题，通过对少数类样本进行数据增强和多数类样本重采样的方法重新构建新的训练数据集，分别提取时域、频域、小波域的特征以表征信号。针对样本不均衡问题，提出了基于新采样数据集的集成学习方案，减少样本不均衡影响同时提高分类准确率。该模型能够在保证准确率的同时保证实时性要求，实验结果显示识别准确率可达99.99%，测试速度为每个样本0.67ms。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	王雪晴1
	2
	3
	刘小军1
	2
	刘艳4
	程璐1
	2
	3
	许鑫1
	2
	纪奕才1
	2
	方广有1
	2

关键词 ：集成学习, 类别不均衡, 小样本学习, 核爆电磁脉冲, 闪电电磁脉冲

Abstract：The classification of nuclear and lightning electromagnetic pulse is a key problem in nuclear identification system, in which the main challenge is that the level of class imbalance can be as huge as . For this purpose, we propose SMALLBAG, a novel ensemble learning method. First, we develop a preprocessing procedure to rebuild training set, in which data augmentation methods are used to generate more minority class samples and resampling scheme is proposed to approach class balance. Second, feature extraction is performed in time, frequency and wavelet domains, which are used to characterize the signal. Finally, the ensemble learning method is proposed to alleviate the influence of class imbalance and improve the performance of identification. Experimental results indicate that the simplicity of the proposed learning method ensures the identification accuracy and real-time requirements at the same time, i.e. 99.99% identification accuracy and 0.67 ms testing time per sample.

Key words： ensemble learning class imbalance small samples nuclear electromagnetic pulse (NEMP) lightning electromagnetic pulse (LEMP)

收稿日期: 2022-06-08 出版日期: 2023-06-15

引用本文:

王雪晴1,2,3，刘小军1,2,刘艳4,程璐1,2,3,许鑫1,2,纪奕才1,2，方广有1,2. 采用最优集成学习的小样本电磁脉冲信号分类[J]. 振动与冲击, 2023, 42(11): 193-198.
WANG Xueqing1,2,3, LIU Xiaojun1,2, LIU Yan4, CHENG Lu1,2,3, XU Xin1,2, JI Yicai1,2, FANG Guangyou1,2. Small samples electromagnetic pulse classification using an optimal ensemble learning method. JOURNAL OF VIBRATION AND SHOCK, 2023, 42(11): 193-198.

链接本文:

http://jvs.sjtu.edu.cn/CN/ 或 http://jvs.sjtu.edu.cn/CN/Y2023/V42/I11/193

[1] Xie Y, Wang Z, Wang Q, et al. High altitude nuclear electromagnetic pulse waveform standards: a review[J]. High Power Laser and Particle Beams, 2003, 15(8): 781-787.
[2] 祁树锋;李夕海;韩绍卿; 等. 基于多重分形分析的核爆与雷电电磁脉冲识别[J]. , 2013, 32(7): 8-10.
QI Shu-Feng; LI Xi-Hai; Han SHAO-Qing; et al. Discrimination of Nuclear Explosion and Lightning Electromagnetic Pulse Using Multi-fractals Analysis. , 2013, 32(7): 8-10.
[3] Qi S F, Li X H, Han S Q, et al. Discrimination of nuclear explosion and lightning electromagnetic pulses using multi-fractal analysis[J]. Zhendong yu Chongji(Journal of Vibration and Shock), 2013, 32(7).
[4] Rakov V A, Uman M A. Lightning: physics and effects[M]. Cambridge university press, 2003.
[5] Heavner M J, Suszcynsky D M, Smith D A. LF/VLF intracloud waveform classification[C]//International Conference on Atmospheric Electricity. Versailles, France. 2003.
[6] Betz H D, Schmidt K, Oettinger P, et al. Lightning detection with 3‐D discrimination of intracloud and cloud‐to‐ground discharges[J]. Geophysical research letters, 2004, 31(11).
[7] Liu F, Zhu B, Lu G, et al. Observations of blue discharges associated with negative narrow bipolar events in active deep convection[J]. Geophysical Research Letters, 2018, 45(6): 2842-2851.
[8] Schmitter E D. Remote sensing and modeling of lightning caused long recovery events within the lower ionosphere using VLF/LF radio wave propagation[J]. Advances in Radio Science, 2014, 12: 241-250.
[9] 祁树锋，李夕海，韩绍卿, 等. 核爆与雷电电磁脉冲识别 [J] . , 强激光与粒子束 , 2012,24(12):2935-2940.
QI Shu-feng, LI Xi-hai, HAN Shao-qing, et al. Discrimination of nuclear explosion and lightning electromagnetic pulse [J]. High Power Laser and Particle Beams,2012,24(12):2935-2940.
[10] 王涛，陈福贵，刘明，等 . 基于时域特征分布的核爆电磁脉冲识别算法[J]. 振动与冲击 ,2020,39(08):159 164.
WANG Tao, CHEN Fu-gui, LIU Ming, et al. A discrimination algorithm of nuclear electromagnetic p ulse based on characteristic parameters in time domain[J]. Journal of Vibration and Shock, 2020,39(08):159 164.
[11] Ballarotti M G, Saba M M F, Pinto Jr O. High‐speed camera observations of negative ground flashes on a millisecond‐scale[J]. Geophysical Research Letters, 2005, 32(23).
[12] Nag A, Murphy M J, Cummins K L, et al. Recent evolution of the us National lightning detection network[C]//23rd International Lightning Detection Conference & 5th International Lightning Meteorology Conference. 2014.
[13] Tsurushima D, Honma N, Tsuchiya F, et al. Matching Algorithms of ELF-LEMPs and Lightning Geo-location Data[J]. IEEJ Transactions on Power and Energy, 2018, 138(5): 339-345.
[14] 韩绍卿;宋仔标;伍海军. 改进的波形复杂度算法在核爆炸监测中的应用 [J]. , 2011, 30(2): 205-209.
HAN Shao-qing; SONG Zi-biao;WU Hai-jun. The application of improved algorithms of P-wave complexity in nuclear explosion monitoring. , 2011, 30(2): 205-209.
[15] Li X, Liu Z, Wu H, et al. On the application of fractal analysis and wavelet in NEMP signal processing[J]. Nuclear Electronics and Detection Technology, 2004, 24(2): 143-146.
[16] Jacobson A R, Knox S O, Franz R, et al. FORTE observations of lightning radio-frequency signatures: Capabilities and basic results[J]. Radio Science, 1999, 34(2): 337-354.
[17] Heavner M J, Smith D A, Harlin J. Current Los Alamos sferic array studies[R]. Los Alamos National Lab.(LANL), Los Alamos, NM (United States), 2000.
[18] 张旭荣 , 张妙兰，刘新中 . 小波变换在核爆电磁脉冲信号识别中的应用 [J]. 电子与信息学报 , 1999, 21(5):710-712.
ZHANG Xu-rong, ZHANG Miao-lan, LIU Xin-zhong. Studies of recognition methods of nuclear and ligh tning pulse signals with applications of wavelet transform[J]. Journal of Electronics, 1999,21(5):710 712.
[19] 张仲山, 张恩山, 高春霞. 小波分析在核爆与闪电识别中的应用[J]. 电波科学学报, 2000, 15(4):5.
ZHANG Zhong-shan, ZHANG En-shan, GAO Chun-xia. The application of wavelet analysis to recognition of nuclear explosion and lightning[J]. CHINESE JOURNAL OF RADIO SCIENCE, 2000, 15(4):5.
[20] Li C, Feng D. Feature extraction of the NEMP waveforms based on the wavelet, fractal and genetic algorithm[J].
[21] 王浩骅，苗家友，朱万华，等 . 神经网络识别算法去除核爆电磁脉冲探测闪电干扰 [J]. 核电子学与探测技术 ,2021,41(01):92-97.
WANG Hao-hua, MIAO Jia-you, ZHU Wan-hua, et al. Neural Network Identification Algorithm to Remove Lightning Interference from Nuclear Explosion EMP Detection[J]. Nuclear Electronics and Detection Technology, 2021,41(01): 92-97.
[22] Eads D R, Hill D, Davis S, et al. Genetic algorithms and support vector machines for time series classification[C]//Applications and Science of Neural Networks, Fuzzy Systems, and Evolutionary Computation V. International Society for Optics and Photonics, 2002, 4787: 74-85.
[23] Li P, Song L, Han C, et al. Recognition of NEMP and LEMP signals based on auto-regression model and artificial neutral network[J]. High Power Laser and Particle Beams, 2010, 22(12): 3052-3056.
[24] Li J, Si Y, Xu T, et al. Deep convolutional neural network based ECG classification system using information fusion and one-hot encoding techniques[J]. Mathematical problems in engineering, 2018, 2018.
[25] Aytar Y, Vondrick C, Torralba A. Soundnet: Learning sound representations from unlabeled video[J]. Advances in neural information processing systems, 2016, 29.
[26] Liu Y, Yao X. Ensemble learning via negative correlation[J]. Neural networks, 1999, 12(10): 1399-1404.
[27] Han H, Wang W Y, Mao B H. Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning[C]//International conference on intelligent computing. Springer, Berlin, Heidelberg, 2005: 878-887.
[28] Chawla N V, Bowyer K W, Hall L O, et al. SMOTE: synthetic minority over-sampling technique[J]. Journal of artificial intelligence research, 2002, 16: 321-357.
[29] He H, Bai Y, Garcia E A, et al. ADASYN: Adaptive synthetic sampling approach for imbalanced learning[C]//2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence). IEEE, 2008: 1322-1328.
[30] Ditterrich T G. Machine learning research: four current direction[J]. Artificial intelligence magzine, 1997, 4: 97-136.
[31] Houghten R A, Harvey R B. Accuracy and Reliability of a Short-Baseline Narol System[R]. AIR FORCE CAMBRIDGE RESEARCH CENTER BEDFORD MA ELECTRONICS RESEARCH DIRECTORATE, 1957.
[32] LIU F F, QIN Z L, ZHU B Y, et al. Observations of ionospheric D layer fluctuations during sunrise and sunset by using time domain waveforms of lightning narrow bipolar events[J]. Chinese journal of geophysics, 2018, 61(2): 484-493.
[33] QIN Z L, ZHU B Y, MA M, et al. Using time domain waveforms of return strokes to retrieve the daytime fluctuation of ionospheric D layer[J]. Chinese Science Bulletin, 2015, 60(7): 654-663.
[34] Chen T, He T, Benesty M, et al. Xgboost: Extreme Gradient Boosting, R Package Version 0.4-2; 2015[J]. 2016.
[35] Chang C C, Lin C J. LIBSVM: a library for support vector machines[J]. ACM transactions on intelligent systems and technology (TIST), 2011, 2(3): 1-27.
[36] Kingma D P, Welling M. Auto-encoding variational bayes[J]. arXiv preprint arXiv:1312.6114, 2013.
[37] Safavian S R, Landgrebe D. A survey of decision tree classifier methodology[J]. IEEE transactions on systems, man, and cybernetics, 1991, 21(3): 660-674.
[38] Prokhorenkova L, Gusev G, Vorobev A, et al. CatBoost: unbiased boosting with categorical features[J]. Advances in neural information processing systems, 2018, 31.
[39] Chen T, Guestrin C. Xgboost: A scalable tree boosting system[C]//Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016: 785-794.
[40] Wang H, Yang Y, Wang H, et al. Soft-voting clustering ensemble[C]//International Workshop on Multiple Classifier Systems. Springer, Berlin, Heidelberg, 2013: 307-318.
[41] LeCun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[42] Lin T Y, Goyal P, Girshick R, et al. Focal loss for dense object detection[C]//Proceedings of the IEEE international conference on computer vision. 2017: 2980-2988.
[43] Kingma D P, Ba J. Adam: A method for stochastic optimization[J]. arXiv preprint arXiv:1412.6980, 2014.