多焦ERG图形刺激器的设计与实现

多焦视网膜电图(mfERG)检测,可以在相对较短的时间内连续记录测试野多个局部视网膜电生理响应信号的波形,有着重要的临床应用价值。多焦ERG图形刺激器作为多焦ERG检测系统的关键部分,充分体现了多焦ERG的记录原理,对视觉电信号的检测起重要作用。对多焦ERG图形刺激器的各种参数及m-序列对图象的控制方式作了论述,通过VC++环境下绘图的基本知识及绘图变换算法,给出了具体的多焦ERG图形刺激器生成算法及实现。     

基于视网膜电图研究的三值序列的产生和分析

多焦视网膜电图在临床眼科疾病诊断中已经被广泛应用,目前多焦视网膜电图主要通过二值m序列来控制黑白图形翻转,由于人眼视网膜神经对光颜色感知度不同,为了获取更多的疾病信息,提出一种新颖的三值m序列并探讨其应用到多焦视网膜电图中的可行性方法。采用三值m序列的寄存器产生原理,在Matlab中对基本性质进行仿真分析。仿真结果表明所提出三值m序列具备了良好的相关性。可以应用到多焦视网膜电图多色刺激中实现对各个刺激斑响应信号的提取。     

基于ICA方法和遗传算法的胎儿心电信号提取

现有独立分量分析(ICA)方法易陷入局部极优值,从而使提取出的胎儿心电信号(FECG)中常混有部分母体心电(MECG).为解决该问题,本文提出一种基于独立分量分析和遗传算法(GA)的新的胎儿心电信号提取方法.该方法首先利用遗传算法在全局空间内搜索最优值,然后又通过基于峭度的固定点算法的迭代核加强局部搜索,因而增强了算法的分离能力.该算法适用于超高斯和亚高斯信号同时存在的情况,并且具有遗传算法所有的收敛性强的优点.实验结果表明,该算法可以成功地分离混叠信号,而且与传统的独立分量分析方法相比,具有更优异的分离能力,提取出的胎儿心电信号噪声小,几乎不混有母体心电.     

小波变换结合盲源分离的EEG情感识别

由于脑电信号独立源数目的不确定性以及其他噪声的干扰,使得采集的脑电信号各导信号之间产生串扰、源信号难以估计以及噪声混杂等问题,严重影响了对脑电信号的分析研究。将小波变换与盲源分离算法相结合,并对盲源分离算法中维格纳分布存在的交叉项干扰现象进行重排处理。主要思路是首先将每一导信号进行小波变换,提取出特征波β波,然后对这些β波信号进行基于重排光滑伪维格纳分布的盲源分离,分离出关联性极大的β波成分。实验结果表明,所用方法分离出了各导信号中关联性大的脑电信号成分,并在一定程度上解决了源信号难以估计等问题,使识别结果有明显的提升。     

基于相关性评估与FastICA的实时心电信号提取算法

针对传统的独立成分分析(ICA)算法无法实现实时的信号分离和提取的问题,提出一种基于相关性评估和实时FastICA的方法.通过对混合信号分段,并进行FastICA处理,再对分离后的信号段进行相关性评估,选取与心电信号相关性最大的信号段进行相位的修正与幅值缩放,实现心电(ECG)信号的实时提取.在分离处理前,对混合信号进行基线漂移的过滤,提高了算法的鲁棒性.实验表明:所提算法的准确率可达96.7％,能够有效实现心电信号的实时提取.     

基于新型能量算子和多尺度熵的癫痫自动检测

提出癫痫检测的新算法,创新地使用新型能量算子和多尺度熵,不仅能够实时跟踪脑电信号能量,而且可以在脑电信号未知情况下提取其有效的特征信息。对脑电信号进行一定尺度的小波变换;利用新型Teager能量算子和多尺度熵提取脑电信号的非线性特征,将特征信号输入到极限学习机中,用来识别正常脑电信号与癫痫脑电信号。实验结果表明,此方法具有高分类精度,可达到98.89%,优于文献中所报道的多种检测方法,可以应用于癫痫临床检测。     

面部局部特征的独立分量分析及其应用

特征提取方法一直是人脸识别研究中的热点,局部特征分析（Local Feature Analysis）算法不仅能得到面部的全局特征,而且能提取出其局部特征信息,但该算法得到的结果具有过多冗余相关信息不利于识别。由于独立成分分析（Independent Component Analysis）算法能够有效地提取信号的高阶统计特性,很好地去除了各分量之间的相关性。给出了融合这两种方法的特征提取方法,经实验测试表明该算法能有效地提取面部特征。     

基于ICA-WICA方法的脑电信号α波提取的仿真研究

脑电信号的波形提取和特征提取成为当前研究的热点,其中较多的是对能反应大脑某种状态波的特征提取,比如反应闭眼脑功能的α波。通常测得的脑电信号会有很多的干扰,用ICA的方法对脑电信号的多种噪声信号进行分离和去除,然后对脑电信号α波成分进行了基于小波变换的多尺度分析,结合ICA的方法对脑电信号α波成分进行二次提取(即本文中提出的ICA-WICA方法)。最后的仿真结果证明了这种方法的有效性。     

基于独立成分分析的表面肌电信号工频去噪

表面肌电信号(SEMG)采集中,如何消除工频干扰对信号的后续应用意义重大.在探讨独立成分分析(ICA)原理的基础上,提出了一种用于表面肌电信号工频去噪的快速独立成分分析(FastICA)算法.该方法通过对观测信号去均值和白化处理后,用负熵作判据通过迭代得到解混矩阵,经解混运算得到源信号.针对混合信号ICA分离效果的差异,引入最大似然指标作为分离效果的评价量.实验结果表明,所提算法能有效分离SEMG信号中的工频噪声,运用最大似然评价指标将工频噪声降至最低.     

基于K近邻互信息估计的EEG伪迹消除方法

针对脑电信号中的眼电和心电串扰伪迹,提出一种基于最小相依成分分析的互信息(MILCA)算法的伪迹消除方法.在提升小波硬阈值法对多路原始脑电信号去噪基础上,运用MILCA算法对各通道信号进行盲源分离,同时采用信号间互相关系数和互信息量作为指标,分析伪迹分离程度.与Extend Infomax、FastICA 2种常见盲源分离算法的对比结果表明,运用MILCA算法对脑电信号中的眼电及心电伪迹的分离结果最理想.     

基于IMF能量谱的水声信号特征提取与分类

经验模态分解（EMD）是用来处理非平稳时变信号的一种信号分析方法,该方法对所分析信号的局部特征信号进行不同时间尺度的分解,从而得到这些局部特征信号的各阶本征模函数（IMF）。提出了一种基于IMF能量谱的水声信号特征提取与选择方法,通过对水声信号进行经验模态分解,提取信号的本征模式分量并转换为能量谱特征向量,从而观测不同信号子频带能量谱的特征变化。分类实验采用支持向量机（SVM）分类器进行。实验结果表明,相对于小波能量谱特征提取法而言,利用IMF能量谱作为特征向量的分类实验具有更佳的分类效果,平均正确率达88%以上。     

Hierarchical projection regression for online estimation of elbow joint angle using EMG signals

Exoskeletons offer great facilities to the elderly and disabled people with respect to extending their moving ranges and reacting to certain physical activities. Electromyogram (EMG) signals, which are derived from the neuromuscular system, provide an important access to the human-robot interface. On one hand, EMG signals can be used for real-time estimation of the motion intention of human body, e.g., the current joint angle status. On the other hand, however, the process of the mass EMG data, which are captured instantaneously from skin surface, challenges the state-of-the-art technology. Because its non-stationary and randomness, it is difficult to extract the valuable and stable features from the raw EMG signals. This paper investigates into the learning process of high dimensional EMG signals with a hierarchical mechanism that projects the original data into a lower feature space to achieve a local refined mapping from the EMG signals to the motion states of the human body. This hierarchically projected regression algorithm constructs incrementally a tree-based knowledge library, whose components represent local regression models. The components will be retrieved online efficiently and contribute to the estimation of the motion states. A great number of experiments are carried out to evaluate the accuracy of this novel algorithm.     

Efficient encoding of natural optic flow

Statistically efficient processing schemes focus the resources of a signal processing system on the range of statistically probable signals. Relying on the statistical properties of retinal motion signals during ego-motion we propose a nonlinear processing scheme for retinal flow. It maximizes the mutual information between the visual input and its neural representation, and distributes the processing load uniformly over the neural resources. We derive predictions for the receptive fields of motion sensitive neurons in the velocity space. The properties of the receptive fields are tightly connected to their position in the visual field, and to their preferred retinal velocity. The velocity tuning properties show characteristics of properties of neurons in the motion processing pathway of the primate brain.     

LMD方法在脑电信号处理中的应用研究

局部均值分解(LMD)是一种新的非线性非平稳信号处理方法,该方法具有较强的自适应性,能将复杂信号分解为一系列具有物理意义的PF(production function)分量。但在信号分解过程中会产生端点效应,这将影响LMD分解精度,严重时会导致信号失真。在详细分析了LMD产生端点效应的原因之后,提出了一种基于相似波形加权匹配的端点延拓算法。通过对视觉诱发脑电信号进行仿真,并且和不做延拓的LMD分解结果做对比,说明该方法能够有效抑制LMD的端点效应,取得较好的分解效果。     

New behavioral‐level simulation technique for RF/microwave applications. Part III: Advanced concepts

The quadrature modeling structure is widely accepted as an efficient tool for the nonlinear simulation of RF/microwave bandpass stages (power amplifiers, etc.) for wireless applications. The common belief is that this structure can be applied to model only bandpass memoryless nonlinearities (which, however, may exhibit amplitude‐to‐phase conversion). In two recent articles 1 , 2 ; the authors have extended the application of the quadrature modeling structure to modeling broadband nonlinearities, which makes possible to predict harmonics and even‐order nonlinearities, to take into account the frequency response, etc. This article completes the overview of the instantaneous quadrature technique. The authors discuss its application to modeling AM, FM and PM detectors, which are strongly nonlinear elements with large memory (both the strong nonlinearity and large memory effects are essential for the detector proper operation), thus removing the limitation of nonlinearity to be memoryless or quasimemoryless. The identification of nonlinear interference/distortion sources is of great relevance for a practical EMC/EMI design. In the second part of this article, we discuss the dichotomous identification method, which is much more computationally efficient than a simple single‐signal method, especially for a large number of input signals. Individual spectral components of a complex‐spectrum signal can also be considered as input signals and, hence, it is possible to identify the spectral components responsible for a particular nonlinear interference/distortion (say, for a particular intermodulation product). © 2002 Wiley Periodicals, Inc. Int J RF and Microwave CAE 12: 206–216, 2002.     

基于替代数据检测的音频信号非线性分析

针对轻微含噪的音频信号,本文提出了一种基于替代数据检测的非线性分析方法。该方法首先根据音频信号的线性假设生成多组替代数据,然后分别计算原始数据和替代数据的样本峰度,最终根据假设检验方法判断原始音频中是否包含非线性成分。实验结果表明,所提方法能够有效地验证音频信号的非线性特性。相比于传统基于最大Lyapunov指数的非线性分析方法,该方法能更好地区分音频信号和噪声信号。     

A dynamical trajectory-based methodology for systematically computing multiple optimal solutions of general nonlinear programming problems

In this paper, a novel dynamical trajectory-based methodology is developed for systematically computing multiple local optimal solutions of general nonlinear programming problems with disconnected feasible components satisfying nonlinear equality/inequality constraints. The proposed methodology, deterministic in nature, exploits trajectories of two different nonlinear dynamical systems to find multiple local optimal solutions. The methodology consists of two phases: Phase I starts from an arbitrary (infeasible) initial point and finds systematically multiple or all the disconnected feasible components; Phase II finds an adjacent local optimal solution from a local optimum via a decomposition point, thereby systematically locating multiple local optimal solutions which lie within each feasible component found in Phase I. By alternating between these two phases, the methodology locates multiple or all the local optimal solutions which lie in all the disconnected feasible components. A theoretical foundation for the proposed methodology is also developed. The methodology is illustrated with a numerical example with promising results.     

基于局部投影和提升小波的混沌信号降噪

根据信号和噪声的不同特性,提出了一种基于局部投影和提升小波的混沌信号降噪方法。该方法利用提升小波变换得到实际混沌信号的近似系数,对其进行相空间重构,与实际混沌信号的重构相空间进行比较,来选取局部投影的邻域,在局部邻域内进行奇异谱分析,利用代表吸引子的主分量来重构混沌信号。通过对Lorenz模型和月太阳黑子进行仿真分析,证实了提出方法对混沌信号降噪的有效性。     

基于条件局部均值分解与变量预测模型的轴承故障诊断方法

针对局部均值分解(LMD)方法在分解非线性、非平稳振动信号过程中存在的模态混淆现象,从而影响故障识别准确性的问题,提出了基于条件局部均值分解方法(CLMD)与模式识别变量预测模型(VPMCD)的故障诊断方法。该方法将数字图像处理的频率分辨率方法与LMD相结合,首先确定振动信号中所有局部极值点的频率分辨率,将振动信号分为低频率分辨率区域和高频率分辨率区域;然后对高频率分辨率区域进行LMD分解,可得若干乘积函数(PF)分量;最后用折线将所有PF分量连接起来,经滑动平均处理可得PF分量,提取PF分量的偏度系数和能量系数构成故障特征向量,用于VPMCD故障识别。将该方法应用于轴承故障诊断,实验结果表明,与LMD方法相比,识别效率提高了8.33%,表明了该方法的有效性和可行性。