融合小波包和神经网络的脑电信号处理方法

doi:10.13190/j.jbupt.2020-208

北京邮电大学学报 ›› 2021, Vol. 44 ›› Issue (3): 94-99.doi: 10.13190/j.jbupt.2020-208

融合小波包和神经网络的脑电信号处理方法

李端玲¹, 成苈委¹, 于功敬², 张忠海², 于淑月²

1. 北京邮电大学现代邮政学院(自动化学院), 北京 100876;
2. 北京航天测控技术有限公司, 北京 100041

收稿日期:2020-10-12 出版日期:2021-06-28 发布日期:2021-06-23
通讯作者: 成苈委(1989-),男,博士生,E-mail:liwei_cheng89@163.com. E-mail:liwei_cheng89@163.com
作者简介:李端玲(1974-),女,教授,博士生导师.
基金资助:
国家自然科学基金项目（51775052）；北京市自然科学基金项目（3212009）

An Electroencephalogram Signal Processing Method Fusing Wavelet Packet and Neural Network

LI Duan-ling¹, CHENG Li-wei¹, YU Gong-jing², ZHANG Zhong-hai², YU Shu-yue²

1. School of Modern Post(School of Automation), Beijing University of Posts and Telecommunications, Beijing 100876, China;
2. Beijing Aerospace Measurement & Control Technology Company Limited, Beijing 100041, China

Received:2020-10-12 Online:2021-06-28 Published:2021-06-23

摘要/Abstract

摘要： 针对运动想象脑电信号处理中分类准确率较低的问题，提出了一种基于能量（二阶矩）小波包变换和莱文伯格-马夸特神经网络算法相结合的运动想象脑电信号处理方法.首先，利用能量方法对信号进行时域分析，选取有效的时序段；然后，使用小波包变换对所选有效时域段的各导信号进行时频分解，选取与想象任务相关的频段信息重构脑电信号特征；最后，将各导信号重构的特征串接，导入基于莱文伯格-马夸特训练算法的神经网络实现最终的任务分类.利用2个脑电信号标准竞赛数据库进行方法验证，分别取得了95.62%和90.13%的分类准确率.与近期的一些研究成果进行对比，可知该方法具有较好的分类效果.

关键词: 运动想象脑电信号, 二阶矩, 小波包变换, 反向传播神经网络, 莱文伯格-马夸特算法

Abstract: Aiming at the classification accuracy of the motor imagery electroencephalogram in processing is low, a processing method based on the combination of energy (second-order moment) wavelet packet transform and Levenberg-Marquardt neural network is proposed. Firstly, the energy method is used to analyze signal in the time domain, and the effective time sequence is selected. Then, wavelet packet transform is used to decompose the time-frequency of each pilot signal in the selected effective time-domain segment, and the frequency information related to the imagination task is selected to reconstruct the signal characteristics. Finally, the features reconstructed by each guide signal are concatenated and imported into the neural network based on the Levenberg-Marquardt training algorithm to realize the task classification. The method was verified by two kinds of electroencephalogram signal standard competition database, and the classification accuracy is 95.62% and 90.13%, respectively. Compared with some recent research results, this algorithm has a better processing effect.

Key words: motor imagery electroencephalogram, second-order moment, wavelet packet transformation, back propagation neural network, Levenberg-Marquardt algorithm

中图分类号:

TP751.2

李端玲, 成苈委, 于功敬, 张忠海, 于淑月. 融合小波包和神经网络的脑电信号处理方法[J]. 北京邮电大学学报, 2021, 44(3): 94-99.

LI Duan-ling, CHENG Li-wei, YU Gong-jing, ZHANG Zhong-hai, YU Shu-yue. An Electroencephalogram Signal Processing Method Fusing Wavelet Packet and Neural Network[J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(3): 94-99.

参考文献

[1] Vaughan T M, Heetderks W J, Trejo L J, et al. Brain-computer interface technology:a review of the second international meeting[J]. IEEE Transactions on Neural Systems & Rehabilitation Engineering:a Publication of the IEEE Engineering in Medicine & Biology Society, 2003, 11(2):94-109.
[2] Nguyen L B, Nguyen A V, Ling S H, et al. Combining genetic algorithm and Levenberg-Marquardt algorithm in training neural network for hypoglycemia detection using EEG signals[C]//2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). New York:IEEE Press, 2013:5386-5389.
[3] Aydemir, Onder. Combining sub-band power features extracted from different time segments of EEG trials[C]//2017 40th International Conference on Telecommunications and Signal Processing (TSP). New York:IEEE Press, 2017:383-386.
[4] Chatterjee R, Bandyopadhyay T, Sanyal D K, et al. Dimensionality reduction of EEG signal using fuzzy discerni-bility matrix[C]//2017 10th International Conference on Human System Interactions (HSI). New York:IEEE Press, 2017:131-136.
[5] Nguyen P, Tran D, Le T, et al. Multi-sphere support vector data description for brain-computer interface[C]//2012 Fourth International Conference on Communications and Electronics (ICCE). New York:IEEE Press, 2012:318-321.
[6] Saka K, Aydemir Ö, Öztürk M. Classification of EEG signals recorded during right/left hand movement imagery using fast Walsh Hadamard transform based features[C]//2016 39th International Conference on Telecommunications and Signal Processing (TSP). New York:IEEE Press, 2016:413-416.
[7] Guo Xiaojing, Wu Xiaopei. Motor imagery EEG classification based on dynamic ICA mixing matrix[C]//2010 4th International Conference on Bioinformatics and Biomedical Engineering. New York:IEEE Press, 2010:1-4.
[8] Schlögl A, Lugger K, Pfurtscheller G. Using adaptive autoregressive parameters for a brain-computer-interface experiment[C]//19th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. ‘Magnificent Milestones and Emerging Opportunities in Medical Engineering’(Cat.No.97CH36136). New York:IEEE Press, 1997:1533-1535.
[9] Cheng Liwei, Li Duanling, Li Xiang, et al. The optimal wavelet basis function selection in feature extraction of motor imagery electroencephalogram based on wavelet packet transformation[J]. IEEE Access, 2019, 7:174465-174481.
[10] Blankertz B. BCI competition Ⅱ final results[EB/OL]. (2003-05-02)[2020-10-02]. http://bbci.de/competition/ii/results/index.html.
[11] Blankertz B. BCI Competition Ⅳ final results[EB/OL]. (2008-12-12)[2020-10-02]. http://bbci.de/competition/iv/results/index.html.
[12] Bentlemsan M, Zemouri E T T, Bouchaffra D, et al. Random forest and filter bank common spatial patterns for EEG-based motor imagery classification[C]//2014 5th International Conference on Intelligent Systems, Modelling and Simulation. New York:IEEE Press, 2015:235-238.
[13] Bustios P, Rosa J L. Restricted exhaustive search for frequency band selection in motor imagery classification[C]//2017 International Joint Conference on Neural Networks (IJCNN). New York:IEEE Press, 2017:4208-4213.
[14] Muthong S, Vateekul P, Sriyudthsak M. Stacked probabilistic regularized LDA on partitioning non-stationary EEG data for left/right hand imagery classification[C]//2016 IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES). New York:IEEE Press, 2016:301-306.
[15] Tian Geliang, Liu Yue. Study on classification of left-right hands motor imagery EEG signals based on CNN[C]//2018 IEEE 17th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC). New York:IEEE Press, 2018:324-329.
[16] Zhang Yu, Nam Chang S, Zhou Guoxu, et al. Temporally constrained sparse group spatial patterns for motor imagery BCI[J]. IEEE Transactions on Cybernetics, 2018, 49(9):3322-3332.

融合小波包和神经网络的脑电信号处理方法

An Electroencephalogram Signal Processing Method Fusing Wavelet Packet and Neural Network

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

[1]	杨丽花, 聂倩, 呼博, 江婷. 联合BP神经网络与基扩展模型的信道预测算法 [J]. 北京邮电大学学报, 2023, 46(3): 13-18.
[2]	易茜, 黎伟, 易树平, 谢俊东. 基于鼠标行为时频联合分析的用户可信认证[J]. 北京邮电大学学报, 2021, 44(4): 121-128.
[3]	赵雄文, 孙宁姚, 耿绥燕, 张钰, 杜飞. 基于最小二乘支持向量机的时变信道建模[J]. 北京邮电大学学报, 2019, 42(5): 29-35.
[4]	余修武, 刘琴, 李向阳, 张可, 肖人榕. 基于改进蚁群的BP神经网络WSN数据融合算法[J]. 北京邮电大学学报, 2018, 41(4): 91-96.
[5]	刁鸣, 安春莲. 冲击噪声背景下的DOA估计新方法[J]. 北京邮电大学学报, 2013, 36(5): 99-104.
[6]	于艳华,宋梅,潘阳发,宋俊德. 改进的基于支持向量机的网络综合评价策略[J]. 北京邮电大学学报, 2007, 30(6): 85-88.
[7]	张锦1,2, 李光3, 曹伍4, 胡瑞芬1. 基于主成分分析的自动文本分类模型研究[J]. 北京邮电大学学报, 2006, 29(s2): 136-138.
[8]	闫桂霞, 赵尔沅. 基于IIR两通道双正交滤波器组的小波包分析[J]. 北京邮电大学学报, 1999, 22(1): 15-19.