基于Lempel-Ziv复杂度和经验模态分解的癫痫脑电信号的检测方法

癫痫脑电信号是非平稳、非线性的,根据此特性我们提出一个基于Lempel-Ziv复杂度和经验模态分解(EMD)的癫痫脑电信号的检测方法,首先将癫痫脑电信号用EMD分解,再分别计算每阶固有模态函数(IMF)的复杂度,最后将得到的复杂度作为特征进行检测.实验用波恩数据库来评估提出的方法.结果表明,该方法检测准确率可达到95.25%,具有准确率高、适应性强等优点.     

基于弹性变分模态提取的时间相关单光子计数信号去噪

单光子激光雷达的回波信号具有极低的信噪比,有效地消除噪声和提取出回波信号特征是提升单光子激光雷达测距精度的关键,变分模态分解算法需要使用者依据经验确定分解本征模态函数数量,不具有适用性和通用性.为此,本文基于时间相关单光子计数信号特点,提出了在变分模态分解中让信号按照指定频率进行聚类分解的变分约束条件,并采用弹性网回归重构不适定问题的求解模型,提出了弹性变分模态提取算法.实验结果表明,在波段850 nm、平均发射功率为25 nW、背景噪声平均功率为19.51μW的条件下,利用该方法,得到了时间相关单光子计数信号重建精度的均方根误差为1.414 ns.同时在不同的累积时间下,能够稳定且快速地提取出回波信号特征,有效地提高了算法的去噪能力和特征提取的性能.     

基于多类运动想象任务的EEG信号分类研究

针对多类运动想象脑电信号个体差异性强和分类正确率比较低的问题,提出了一种时-空-频域相结合的脑电信号分析方法：首先利用小波包对EEG原始信号进行分解,根据EEG信号的频域分布提取出运动想象脑电节律,通过“一对多”共空间模式（CSP）算法对不同运动想象任务的脑电节律进行空间滤波提取特征;然后将特征向量输入到“一对多”模式下的支持向量机（SVM）中,并利用判断决策函数值的方法对SVM的输出结果进行融合;最后通过引入时间窗对脑电信号进行时域滤波,消除运动想象开始和结束时脑电的波动,进一步提高信号信噪比和算法的分类效果。实验结果显示：在时间窗为2s时,平均最大 系数达到了0.72,比脑机接口竞赛第一名提高了0.15,验证了该算法能够有效减小脑电信号个体差异性影响,提高多类识别正确率。     

基于AdaBoost算法的癫痫脑电信号识别

AdaBoost算法作为Boosting算法的经典算法之一, 在人脸检测和目标跟踪等领域得到了广泛应用, 但该算法也有一个缺点-退化问题. 为了解决这个问题, 通过对弱分类器进行筛选、引入平滑因子和权值修正函数三个措施对算法进行优化, 并将优化后的算法与小波包分解相结合应用到癫痫脑电信号的识别上. 结果表明, 本文算法对癫痫脑电信号的识别率为96.11%, 对正常脑电信号的识别率为99.51%, 具有较高的识别率, 为癫痫的正确诊断提供了一种可能有效的解决方案.     

基于变分模态分解-传递熵的脑肌电信号耦合分析

皮层肌肉功能耦合是大脑皮层和肌肉组织间的相互作用, 脑肌电信号的多尺度耦合特征可以体现皮层-肌肉间多时空的功能联系. 本文引入变分模态分解并与传递熵结合, 构建变分模态分解-传递熵模型应用于脑肌间耦合研究. 首先基于变分模态分解将同步采集的脑电(EEG) 和肌电(EMG) 信号分别进行时频尺度化, 然后计算不同时频尺度间的传递熵值, 获取不同耦合方向(EEG→EMG 及EMG→EEG) 上不同尺度间的非线性耦合特征. 结果表明, 在静态握力输出条件下, 皮层与肌肉beta (15—35 Hz) 频段间的耦合强度最为显著; EEG→EMG 方向上脑电与肌电高gamma (50—72 Hz) 频段的耦合强度总体上高于EMG→EEG 方向.研究结果揭示皮层-肌肉功能耦合具有双向性, 且脑肌间不同耦合方向上、不同频段间的耦合强度有所差异.因此可利用变分模态分解-传递熵方法定量刻画大脑皮层与肌肉各时频段之间的非线性同步特征及功能联系.     

基于改进变分模态分解的北极海域声速剖面分类*

应用支持向量机对北极声速剖面进行分类,特征量提取是关键。该文采用一种基于经验模态分解的改进变分模态分解算法,以准确提取声速剖面特征量。算法首先对声速剖面信号进行经验模态分解,依据最大类间方差原则划分各分量边际谱主频带,以相似度作为最小分解层数判断标准,获得最小分解层数,进行变分模态分解。对北极区海水声速实测数据(信号)处理表明,该方法可有效提取信号经验模态分解各分量的希尔伯特边际谱特征,进行支持向量机分类,实现对北极海域声速剖面的分类识别,解决以往人工分类耗时久的问题。     

基于频率切片小波变换和支持向量机的癫痫脑电信号自动检测

实现癫痫脑电信号的自动检测对癫痫的临床诊断和治疗具有重要意义.本文提出先使用频率切片小波变换分离出5个不同频段的节律信号,再分别计算每个节律信号的近似熵和相邻节律的波动指数,最后使用遗传算法优化的支持向量机进行分类.实验结果表明,所提出的方法能够对正常、癫痫发作间期和癫痫发作期三种脑电信号进行准确分类,分类准确率为98.33%.     

基于变分模态分解与多尺度排列熵的生物组织变性识别

根据高强度聚焦超声(HIFU)治疗中超声散射回波信号的特点,本文利用变分模态分解(VMD)与多尺度排列熵(MPE)对生物组织变性识别进行了研究.首先对生物组织中的超声散射回波信号进行变分模态分解,根据各阶模态的功率谱信息熵值分离出噪声分量和有用分量;对分离出的有用信号进行重构并提取其多尺度排列熵;然后通过Gustafson-Kessel (GK)模糊聚类确定聚类中心,采用欧氏贴近度与择近原则对生物组织进行变性识别.将所提方法应用于HIFU治疗中超声散射回波信号实验数据,用遗传算法对多尺度排列熵的参数优化后,对293例未变性组织和变性组织的超声散射回波信号数据进行了多尺度排列熵分析,发现变性组织的超声散射回波信号的多尺度排列熵值要高于未变性组织;多尺度排列熵可以较好地识别生物组织是否变性.相对于EMD-MPE-GK模糊聚类以及VMD-小波熵(WE)-GK模糊聚类变性识别方法,本文所提方法中变性与未变性组织特征交叠区域数据点更少,聚类效果和分类性能更好;本实验环境下生物组织变性识别结果表明,该方法的识别率更高,高达93.81%.     

多类运动想象脑电信号特征提取与分类

针对多类运动想象情况下存在的脑电信号识别正确率比较低的问题,提出了一种将小波包方差,小波包熵和共同空间模式相结合的脑电信号特征提取,输入到支持向量机达到分类目的。首先选择小波包去噪后重要导联的脑电信号,进行小波包分解;然后对通道优化选取的重要导联的每个通道信号计算方差和熵值,对重要导联的每个通道信号的子带系数进行重构后,进行共同空间模式特征提取;最后结合2种不同导联方式所获取的特征向量进行分类。采用BCI2005desc_IIIa中l1b数据,该算法的分类正确率最高达到88.75%,相对2种单一的提取方法分别提高28.27%和6.55%。结果表明该算法能够有效提取特征向量,进而改善多类识别正确率较低的问题。     

变分模态分解-排列熵方法用于分布式光纤振动传感系统去噪

提出一种变分模态分解-排列熵的去噪方法,分析并设定排列熵中关键参数和阈值,进而通过排列熵来确定变分模态分解的分解层数值,将分解的各模态进行重构以实现对振动信号的去噪。通过仿真测试来验证该方法在正交性、完备性、信噪比和效率方面的优越性,最后对系统采集的实际振动信号进行去噪处理。实验结果表明,与现有的经验模态分解-相关系数和完全经验模态分解-相关系数方法相比,所提方法对触网、车轮碾压和雨淋三种振动信号具有最优的去噪信噪比(含噪信号与降噪值之比),分别为32.5358 dB、30.5546 dB和29.3435 dB,耗时也较少,分别为1.4432,1.6320,1.2349 s,信号模式识别准确率最高,均在99%以上。     

Multiscale Permutation Lempel–Ziv Complexity Measure for Biomedical Signal Analysis: Interpretation and Application to Focal EEG Signals

This paper analyses the complexity of electroencephalogram (EEG) signals in different temporal scales for the analysis and classification of focal and non-focal EEG signals. Futures from an original multiscale permutation Lempel–Ziv complexity measure (MPLZC) were obtained. MPLZC measure combines a multiscale structure, ordinal analysis, and permutation Lempel–Ziv complexity for quantifying the dynamic changes of an electroencephalogram (EEG). We also show the dependency of MPLZC on several straight-forward signal processing concepts, which appear in biomedical EEG activity via a set of synthetic signals. The main material of the study consists of EEG signals, which were obtained from the Bern-Barcelona EEG database. The signals were divided into two groups: focal EEG signals (n = 100) and non-focal EEG signals (n = 100); statistical analysis was performed by means of non-parametric Mann–Whitney test. The mean value of MPLZC results in the non-focal group are significantly higher than those in the focal group for scales above 1 (p < 0.05). The result indicates that the non-focal EEG signals are more complex. MPLZC feature sets are used for the least squares support vector machine (LS-SVM) classifier to classify into the focal and non-focal EEG signals. Our experimental results confirmed the usefulness of the MPLZC method for distinguishing focal and non-focal EEG signals with a classification accuracy of 86%.     

X-ray Pulsar Signal Denoising Based on Variational Mode Decomposition

Pulsars, especially X-ray pulsars detectable for small-size detectors, are highly accurate natural clocks suggesting potential applications such as interplanetary navigation control. Due to various complex cosmic background noise, the original pulsar signals, namely photon sequences, observed by detectors have low signal-to-noise ratios (SNRs) that obstruct the practical uses. This paper presents the pulsar denoising strategy developed based on the variational mode decomposition (VMD) approach. It is actually the initial work of our interplanetary navigation control research. The original pulsar signals are decomposed into intrinsic mode functions (IMFs) via VMD, by which the Gaussian noise contaminating the pulsar signals can be attenuated because of the filtering effect during signal decomposition and reconstruction. Comparison experiments based on both simulation and HEASARC-archived X-ray pulsar signals are carried out to validate the effectiveness of the proposed pulsar denoising strategy.     

Rolling Bearing Fault Diagnosis Based on VMD-MPE and PSO-SVM

The goal of the paper is to present a solution to improve the fault detection accuracy of rolling bearings. The method is based on variational mode decomposition (VMD), multiscale permutation entropy (MPE) and the particle swarm optimization-based support vector machine (PSO-SVM). Firstly, the original bearing vibration signal is decomposed into several intrinsic mode functions (IMF) by using the VMD method, and the feature energy ratio (FER) criterion is introduced to reconstruct the bearing vibration signal. Secondly, the multiscale permutation entropy of the reconstructed signal is calculated to construct multidimensional feature vectors. Finally, the constructed multidimensional feature vector is fed into the PSO-SVM classification model for automatic identification of different fault patterns of the rolling bearing. Two experimental cases are adopted to validate the effectiveness of the proposed method. Experimental results show that the proposed method can achieve a higher identification accuracy compared with some similar available methods (e.g., variational mode decomposition-based multiscale sample entropy (VMD-MSE), variational mode decomposition-based multiscale fuzzy entropy (VMD-MFE), empirical mode decomposition-based multiscale permutation entropy (EMD-MPE) and wavelet transform-based multiscale permutation entropy (WT-MPE)).     

Continuous EEG-fMRI in patients with partial epilepsy and focal interictal slow-wave discharges on EEG

Purpose

To verify whether in patients with partial epilepsy and routine electroenecephalogram (EEG) showing focal interictal slow-wave discharges without spikes combined EEG–functional magnetic resonance imaging (fMRI) would localize the corresponding epileptogenic focus, thus providing reliable information on the epileptic source.

Methods

Eight patients with partial epileptic seizures whose routine scalp EEG recordings on presentation showed focal interictal slow-wave activity underwent EEG–fMRI. EEG data were continuously recorded for 24 min (four concatenated sessions) from 18 scalp electrodes, while fMRI scans were simultaneously acquired with a 1.5-Tesla magnetic resonance imaging (MRI) scanner. After recording sessions and MRI artefact removal, EEG data were analyzed offline. We compared blood oxygen level-dependent (BOLD) signal changes on fMRI with EEG recordings obtained at rest and during activation (with and without focal interictal slow-wave discharges).

Results

In all patients, when the EEG tracing showed the onset of focal slow-wave discharges on a few lateralized electrodes, BOLD-fMRI activation in the corresponding brain area significantly increased. We detected significant concordance between focal EEG interictal slow-wave discharges and focal BOLD activation on fMRI. In patients with lesional epilepsy, the epileptogenic area corresponded to the sites of increased focal BOLD signal.

Conclusions

Even in patients with partial epilepsy whose standard EEGs show focal interictal slow-wave discharges without spikes, EEG–fMRI can visualize related focal BOLD activation thus providing useful information for pre-surgical planning.     

遗传算法优化变分模态分解提取舰船辐射噪声特征线谱方法

特征线谱提取是舰船目标识别的一个重要研究环节,常采用传统的DEMON谱分析方法,处理过程中,一般对舰船噪声时域信号未予抑噪,低信噪比情况下,传统DEMON谱分析性能差。对此,提出一种采用遗传算法优化变分模态分解方法,用于分解舰船噪声原时域信号,获得抑制噪声后的舰船噪声重构信号,进而有效提取了舰船目标噪声幅度调制特征线谱。该方法首先采用遗传算法优化变分模态分解的两个关键输入参数(分解所取模态个数和惩罚因子),对变分模态分解得到的各阶固有模态分量加以判别,去除噪声主导分量,保留信号主导分量,使重构舰船噪声信号显著抑制了干扰噪声,然后对降噪后的重构信号进行频谱分析,获得目标噪声调制特征线谱。理论分析、仿真和实验数据处理结果表明,相比传统DEMON谱分析法,基于遗传算法优化变分模态分解的舰船噪声特征线谱提取方法具有更好的噪声抑制能力,所获取的舰船噪声幅度调制特征线谱信噪比明显高于传统DEMON方法,具有一定优势,前景良好。     

基于改进经验模态分解域内心动物理特征识别模式分量的心电信号重建

心电图(electrocardiogram,ECG)诊断心脏疾病的严格标准,要求有效地消除噪声并准确地重建ECG信号.经验模式分解(empirical mode decomposition,EMD)方法重建ECG信号中,模式混叠及重建采用模式分量的识别以经验为基础,导致重建ECG信号准确度降低,且方法不具有自适应和通用性.本文首先基于积分均值定理提出一种改进的EMD方法——积分均值模式分解(integral mean mode decomposition,IMMD)方法,经5000个高斯白噪声样本的蒙特卡罗法验证,IMMD方法比EMD具有更优多分辨率分析能力,能够有效地缓解模式混叠.其次,基于ECG信号内固有心动物理特征量识别重建ECG信号所采用的模式分量,具有现实物理意义,因此,方法具有自适应和通用性.经验证,提出方法重建47例ECG信号与原ECG信号的相关系数中:31例优于变分模式分解方法;33例优于Haar小波软阈值法;42例优于集总经验模式分解方法;45例优于EMD方法.相关系数均值为0.8904,方差为0.0071,表现稳定且最优.     

URLLC occasional large time delay prediction based on unbalanced regression and LSTM

Occasional large time delay is one of the bottlenecks for 5G applied to the industrial field. Accurate measurement, analysis, and prediction of communication network latency are of great significance. Due to the strong randomness and small number of samples in ultra-reliable and low latency communication (URLLC), the prediction task of occasional large time delay is very challenging. This paper proposes an URLLC occasional large time delay prediction method based on unbalanced regression algorithms and long–short term memory (LSTM). We first decompose the original sequence by using the variational mode decomposition (VMD) to obtain relatively stable component sequences. The parameters of the VMD are automatically optimized by grasshopper optimization algorithm (GOA). In order to improve the prediction effect of URLLC occasional large time delay, we introduce unbalanced regression algorithms. Before the VMD decomposition of data samples, the SMOGN is introduced to balance the number of large-time delay samples and common-time delay samples. Next we calculate the weight of each sample based on the rarity of each sample point and our proposed LDSWeight. Finally, LSTM is used to complete cost-sensitive learning. The experimental results show that the URLLC occasional large time delay prediction method proposed in this paper has better prediction accuracy than other methods.     

Research on Driving Fatigue Alleviation Using Interesting Auditory Stimulation Based on VMD-MMSE

The accurate detection and alleviation of driving fatigue are of great significance to traffic safety. In this study, we tried to apply the modified multi-scale entropy (MMSE) approach, based on variational mode decomposition (VMD), to driving fatigue detection. Firstly, the VMD was used to decompose EEG into multiple intrinsic mode functions (IMFs), then the best IMFs and scale factors were selected using the least square method (LSM). Finally, the MMSE features were extracted. Compared with the traditional sample entropy (SampEn), the VMD-MMSE method can identify the characteristics of driving fatigue more effectively. The VMD-MMSE characteristics combined with a subjective questionnaire (SQ) were used to analyze the change trends of driving fatigue under two driving modes: normal driving mode and interesting auditory stimulation mode. The results show that the interesting auditory stimulation method adopted in this paper can effectively relieve driving fatigue. In addition, the interesting auditory stimulation method, which simply involves playing interesting auditory information on the vehicle-mounted player, can effectively relieve driving fatigue. Compared with traditional driving fatigue-relieving methods, such as sleeping and drinking coffee, this interesting auditory stimulation method can relieve fatigue in real-time when the driver is driving normally.     

基于变分模态分解的语音情感识别方法*

针对传统语音情感特征参数在进行情感分类时性能不佳的问题,该文提出了一种基于变分模态分解的语音情感识别方法。情感语音信号首先由变分模态分解提取固有模态函数,然后对所选主导固有模态函数进行重新聚合,再提取梅尔倒谱系数和各固有模态函数的希尔伯特边际谱。为了验证该文提出的特征性能,选用两种语音数据库(EMODB、RAVDESS)进行实验,按该文方法提取特征后使用极限学习机进行语音情感分类识别。实验结果表明:相比基于经验模态分解和集合经验模态分解的语音情感特征,该文提出的特征有更好的识别性能,验证了该方法的实用性。