脉搏的产生来自于心脏的搏动，心脏的收缩舒张使血液通过动脉血管流动到达身体的各个组织器官和身体末梢。因此脉搏信号不仅反应了心脏的工作状态，也包含了身体内部各脏器的生理状况信息，这导致脉搏信号是一个包含众多不确定性因素的复杂信号。本文将复杂性测度引入脉搏信号预处理，特征提取和分类识别中，工作如下：

针对脉搏信号非线性且成分复杂的特点和样本熵对信号分量幅值变化不敏感的局限性，提出一种新的基于信息熵的复杂性测度方法——加权样本熵（Weighted Sample Entropy，WSE）,该算法不仅可以检测出系统的变化，而且对信号分量的振幅变化比其他方法更敏感。

将加权排列熵（Weighted Permutation Entropy，WPE）与变分模态分解（Variational Mode Decomposition，VMD）相结合的方法应用于脉搏信号滤波。该算法通过获取本征模态分量（Intrinsic Mode Function，IMF）中心频率最大方差所对应的分解模态数和二次惩罚项对脉搏信号进行变分模态分解，通过筛选IMF的WPE排除主观选择参数带来的人为误差，实现了对信号的重构，去除了基线漂移和高频噪声。

将复杂性测度引入脉搏信号特征提取中，提出将样本熵、加权样本熵、排列熵、加权排列熵作为脉搏信号的特征。

在麻省理工学院提供的脉搏信号情感数据样本和自建的脉搏情感样本中，对比了加权样本熵、样本熵、排列熵和加权排列熵对不同情感的脉搏信号的区分能力。实验结果显示，本文提出的加权样本熵算法相对于另外三种熵算法，可以有效区分不同情感状态下的脉搏信号。

The pulse is generated from the beating of the heart. The contraction and relaxation of the heart allows blood to flow through the arteries to reach the various tissues and organs of the body and the ends of the body. Therefore, the pulse signal not only reflects the working state of the heart, but also contains information about the physiological conditions of the internal organs of the body. This results in the pulse signal being a complex signal containing many uncertain factors. In this thesis, complexity measurement is introduced into pulse signal preprocessing, feature extraction and classification and recognition. The work is as follows:

Aiming at the nonlinearity and complex composition of the pulse signal and the limitation that the sample entropy is not sensitive to the change of the signal component amplitude, a new complexity measurement method based on information entropy-weighted sample entropy (WSE) is proposed in this thesis. This algorithm not only can detect system changes but also is more sensitive to changes in signal component amplitude than other methods.

The method combining weighted permutation entropy (WPE) and variational mode decomposition (VMD) is applied to pulse signal filtering. The algorithm first performs the variational modal decomposition of the pulse signal by obtaining the decomposition mode number corresponding to the maximum variance of the center frequency of the intrinsic mode component (IMF) and the secondary penalty term, and eliminates the subjective selection parameters by filtering the WPE of the IMF, for human error. the signal is reconstructed to remove baseline drift and high-frequency noise.

The complexity measure is introduced into the feature extraction of pulse signal, the sample entropy, weighted sample entropy (WSE), permutation entropy and weighted permutation entropy (WPE) are proposed as the features of pulse signal.

Finally, the ability of weighted sample entropy is compared with sample entropy, permutation entropy and weighted permutation entropy to distinguish pulse signals of different emotions in the pulse signal emotion data samples provided by MIT and the pulse emotion samples built by ourselves. Experimental results show that the weighted sample entropy algorithm proposed in this thesis can effectively distinguish pulse signals in different emotional states compared with the other three entropy algorithms.

[1] 李雪, 李福凤. 脉象信息分析法的研究进展[J]. 中华中医药志, 2017, 32(10): 4558-4561.

[2] 王礼立, 王晖, 杨黎明, 等. 论脉搏波客观化和定量化研究的症结所在[J]. 中华中医药杂志, 2017, 32(11): 4855-4863.

[3] 姜瑞雪, 马作峰, 王平, 等.《黄帝内经》时间相关的脉诊理论及其应用[J]. 中国中医基础医学杂志, 2016, 22(02): 152-153.

[4] 汪南玥, 于友华, 闪增郁, 等.脉诊信息分析方法[J]. 中国中医基础医学杂志, 2011(11): 1287-1288.

[5] 千英信, 汤建国, 张志枫, 等. 基于寸口脉搏波频谱分析的体外循环血液灌注效果研究[J]. 中国中医基础医学杂志, 2019, 25(04): 481-486.

[6] 徐克, 赵良举, 李明阳. 基于流体网络的人体血液体循环分析[J]. 中国生物医学工程学报, 2017, 36(05): 580-588.

[7] 钱学森, 于景元, 戴汝为. 一个科学新领域——开放的复杂巨系统及其方法论[J]. 自然杂志, 1990(01): 3-10+64.

[8] Soundirarajan M, Aghasian E, Krejcar O, et al. Complexity-based analysis of the coupling between facial muscle and brain activities[J]. Biomedical Signal Processing and Control, 2021, 67(04): 511-517.

[9] Ahamed Mohamed Rasmi Ashfaq,Babini Mohammad Hossein,Namazi Hamidreza. Analysis of the information transfer between brains during a conversation[J]. Technology and Health Care,2021,29(2). 6149-6152.

[10] Ling H, Lech M, Jing Z, et al. Study of wavelet packet energy entropy for emotion classification in speech and glottal signals[J]. Proc Spie, 2013, 8878(4):34.

[11] Wang X W, Nie D, Lu B L. Emotional state classification from EEG data using machine learning approach[J]. Neurocomputing, 2014, 129(apr.10):94-106.

[12] Vyrynen E, Kortelainen J, Seppnen T. Classifier-based learning of nonlinear feature manifold for visualization of emotional speech prosody[J]. IEEE Transactions on Affective Computing, 2013, 4(1):47-56.

[13] 邢广鑫, 周以齐. 基于LMD排列熵和LLE的滚动轴承故障诊断[J]. 煤矿机械, 2021, 42(03): 164-166.

[14] 段萍, 王旭, 丁承君, 等.基于IMF奇异值熵和t-SNE的滚动轴承故障识别[J]. 传感器与微系统, 2021, 40(03): 134-137.

[15] 陈芒, 于德介, 高艺源.基于可视图图谱幅值熵的滚动轴承故障诊断方法[J]. 振动与冲击, 2021, 40(04): 23-29.

[16] 曾莼, 顾荣宝. 奇异值分解熵对股票市场的预测力研究——基于相空间重构技术[J]. 金融经济, 2016(04): 77-79.

[17] ]余海华. 系统复杂性熵理论下股市有效性与复杂性的变化分析——以股权分置改革的视角[J]. 系统科学学报, 2021(02): 58-63.

[18] 杨甲森, 孟新, 王春梅. 卫星遥测数据相关性知识发现方法[J]. 国防科技大学学报, 2019, 41(05): 71-78.

[19] Chung Y F, Hu C S, Yeh C C, et al. How to standardize the pulse-taking method of traditional Chinese medicine pulse diagnosis[J]. Computers in biology and medicine, 2013, 43(4): 342-349.

[20] Nawsupe G, Joshi R R. Modified wavelet-based technique for baseline drift removal and diagnostic scope of spectral energy of radial pulse signal[J]. International Journal of Biomedical Engineering and Technology, 2011, 6(1): 1-13.

[21] 王金海, 岳晨飞, 韦然, 等. 基于EMD的动态脉搏数据处理研究[J]. 仪器仪表学报, 2016, 37(S1): 34-39.

[22] Li-sheng Xu, Yan-hua S, Yue Z, et al. EMD-based Adaptive Wavelet Threshold for Pulse Wave Denoising[J]. Chinese Journal of Biomedical Engineering, 2015, 24(01): 1-8.

[23] 刘攀, 夏春明, 燕海霞, 等. 基于 EEMD 的脉搏信号改进阈值去噪研究[J]. 计算机应用与软件, 2016, 33(3): 76-79.

[24] Yue H A, Li X B, Cai K, et al. Non-contact heart rate detection by combining empirical mode decomposition and permutation entropy under non-cooperative face shake[J]. Neurocomputing, 2020, 392: 142-152.

[25] 张红, 王龙, 李小娜. 一种用CEEMDAN和排列熵去除脉搏信号噪声的方法[J].中国科技论文, 2019, 14(03): 250-254.

[26] Lin G M,B Haryadi,Yang C M, et al. Discrepancies between Conventional Multiscale Entropy and Modified Short-Time Multiscale Entropy of Photoplethysmographic Pulse Signals in Middle- and Old- Aged Individuals with or without Diabetes[J]. Entropy, 2017, 19(3): 1-10.

[27] 张爱华, 常婷婷, 漆宇晟, 等.基于脉搏信号融合分析的心率监测方法[J].电子测量与仪器学报,2020,34(10):163-171.

[28] Danielewska M E, Placek M M, AK Kicińska, et al. Using the entropy of the corneal pulse signal to distinguish healthy eyes from eyes affected by primary open-angle glaucoma[J]. Physiological Measurement, 2020, 41(5): 055-064.

[29] Hosseini, S.A. Naghibi-Sistani, M.-B. Emotion recognition method using entropy analysis of EEG signals[J]. International Journal of Image, Graphics and Signal Processing(IJIGSP),2011,3(5): 30-36.

[30] [1] Xin L, Cai E, Tian Y, et al. An improved electroencephalogram feature extraction algorithm and its application in emotion recognition[J]. Journal of Biomedical Engineering, 2017, 34(4): 510-517.

[31] Mohammadi, Z. Frounchi, J.; Amiri, M. Wavelet-based emotion recognition system using EEG signal[J]. Neural Computing and Applications. 2017, 28(8): 1985-1990.

[32] 刘国华, 周文斌. 基于卷积神经网络的脉搏波时频域特征混叠分类[J]. 吉林大学学报(工学版), 2020, 50(05): 1818-1825.

[33] 张爱华, 靳冠军. 基于脉搏波传导时间变异性的冠心病识别方法[J]. 中国医学物理学杂志, 2017, 34(05): 527-530+540.

[34] Rui G, Wang Y, Yan H, et al. Analysis and Recognition of Traditional Chinese Medicine Pulse Based on the Hilbert-Huang Transform and Random Forest in Patients with Coronary Heart Disease[J]. Evidence-based Complementary and Alternative Medicine, 2015, 2015: 1-8.

[35] 蒋文都, 夏春明, 章悦, 等. 基于肌音信号分形维数的上斜方肌静态疲劳研究[J]. 航天医学与医学工程, 2019, 32(03): 259-264.

[36] 顾玲玲, 张晓俊, 黄程韦, 等.息肉与麻痹喉声源分类中非线性动力学发声系统模型研究[J]. 声学学报, 2015, 40(06): 878-885.

[37] Parlitz U.Estimating Lyapunov exponent from time series [M]. Chao Detection and Predicability. Spring, Berlin, Heidelberg, 2016: 1-34.

[38] Kang H, Zhang X, Zhang G. Phase permutation entropy: A complexity measure for nonlinear time series incorporating phase information[J]. Physica A: Statistical Mechanics and its Applications, 2021, 568: 006-014.

[39] Pincus S M. Approximate entropy as a measure of system complexity[J]. Proceedings of the National Academy of Sciences of the United States of America, 1991, 88(6): 2297-2301.

[40] Richman J S, Randall M J. Physiological time-series analysis, using approximate entropy and sample entropy[J]. American Journal of Physiology-Heart and Circulatory Physiology, 2000, 278(6): 2039-2049.

[41] Bandt C, Pompe B. Permutation entropy: a natural complexity measure for time series[J]. Physical Review Letters, 2002, 88(17): 174-182.

[42] Fadlallah B, Chen B, Keil A, et al. Weighted-permutation entropy: A complexity measure for time series incorporating amplitude information[J]. Physical Review E Statistical Nonlinear & Soft Matter Physics, 2013, 87(2): 022911-1-022911-7.

[43] Jilin University, Patent Issued for Method For Random Noise Reduction From MRS Oscillating Signal Using Joint Algorithms Of EMD And TFPF [J]. Journal of Mathematics,2020,56(07): 375-383.

[44] Wu Z, Huang N E, Chen X. The Multi-Dimensional ensemble empirical mode decomposition method[J]. Advances in Adaptive Data Analysis,2009,1(03): 339-372.

[45] Maria D. van der Walt. Empirical mode decomposition with shape-preserving spline interpolation[J]. Results in Applied Mathematics,2020,5: 17-25.

[46] Yao J, Xiang Y, Qian S, et al. Noise source identification of diesel engine based on variational mode decomposition and robust independent component analysis[J]. Applied Acoustics, 2017, 116: 184-194.

[47] Dragomiretskiy K, Zosso D. Variational Mode Decomposition[J]. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.

[48] 方军强. 基于变分模态分解的船用齿轮箱故障诊断研究[D]. 武汉理工大学, 2017.

[49] 陈凯. 基于变分模态分解的船舶轴系状态监测及故障诊断方法研究[D]. 武汉理工大学, 2018.

[50] Reshef D N, Reshef Y A, Finucane H K, et al. Detecting Novel Associations in Large Data Sets[J]. Science, 2011, 334(6062): 1518-1524.

[51] Xie Dongri, Esmaiel Hamada, Sun Haixin, et al. Feature Extraction of Ship-Radiated Noise Based on Enhanced Variational Mode Decomposition, Normalized Correlation Coefficient and Permutation Entropy.[J]. Entropy (Basel, Switzerland),2020,22(4): 352-360.

[52] 丑永新, 张爱华, 顾亚, 等.一种快速的数学形态学滤波方法及其在脉搏信号处理中的应用[J]. 仪器仪表学报, 2020, 41(02): 253-262.

[53] Cui J, Tu L, Zhang J, et al. Analysis of pulse signals based on array pulse volume[J]. Chinese journal of integrative medicine, 2019, 25(2): 103-107.

[54] 费兆馥. 现代中医脉诊学[M]. 北京:人民卫生出版社,2003: 162-171.

[55] 林炳岐, 关静, 戴宁等. 中医脉诊学现代化研究[J]. 世界中医药, 2017, 12(7): 1706-1710.

[56] 张爱华, 王倩, 丑永新. 基于无线通信的多信号实时采集与分析系统[J]. 控制工程, 2017, 24(04): 875-880.

[57] 鲍凯林, 陈兆学. 基于多尺度样本熵方法分析欧拉视频放大技术在非接触式测量的作用[J]. 中国医学物理学杂志, 2020, 37(005): 635-640.

[58] Nie J, Ji M, Chu Y, et al. Human pulses reveal health conditions by a piezoelectret sensor via the approximate entropy analysis[J]. Nano Energy, 2019, 58: 528-535.

[59] Yu ShunHsin,Chang JenShuo,Tsai ChiaHung Dylan. Grasp to See—Object Classification Using Flexion Glove with Support Vector Machine[J]. Sensors,2021,21(4). 273-297.

[60] Bi L, Xie W, Zhao J. Automatic recognition and classification of multi-channel microseismic waveform based on DCNN and SVM[J]. Computers & geosciences, 2019, 123(FEB.): 111-120.

[61] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2015: 126-128.

[62] Gross J J, Levenson R W . Emotion elicitation using films[J]. Cognition & Emotion, 2007, 9(1): 9-28.