一种心音信号盲源分离方法

为了有效实现单路心音混合信号的盲分离,本文提出了一种基于EMD分解和独立成分函数的单路含噪心音信号盲源分离的方法。首先讨论了单路混合信号的分离模型,含噪信号预处理的方法,以及如何利用EMD变换进行窄带分层和获取独立成分函数的技术;然后通过独立成分函数作为基函数对单路含噪心音信号进行分解,使单路心音信号由一维向量转变为多维向量,从而实现心音信号的盲源分离;最后通过实际的心音分离实验,验证了本方法的实用性,其分离结果的相似度达到0.9792。     

Blind separation of convolutive mixtures of cyclostationary signals

An adaptive blind source separation algorithm for the separation of convolutive mixtures of cyclostationary signals is proposed. The algorithm is derived by applying natural gradient iterative learning to a novel cost function which is defined according to the wide sense cyclostationarity of signals and can be deemed as a new member of the family of natural gradient algorithms for convolutive mixtures. A method based on estimating the cycle frequencies required for practical implementation of the proposed algorithm is presented. The efficiency of the algorithm is supported by simulations, which show that the proposed algorithm has improved performance for the separation of convolved cyclostationary signals in terms of convergence speed and waveform similarity measurement, as compared to the conventional natural gradient algorithm for convolutive mixtures. Copyright © 2004 John Wiley & Sons, Ltd.     

Fixed-point blind source separation algorithm based on ICA

This paper introduces the fixed-point learning algorithm based on independent component analysis (ICA); the model and process of this algorithm and simulation results are presented. Kurtosis was adopted as the estimation rule of independence. The results of the experiment show that compared with the traditional ICA algorithm based on random grads, this algorithm has advantages such as fast convergence and no necessity for any dynamic parameter, etc. The algorithm is a highly efficient and reliable method in blind signal separation. __________ Translated from Journal of Taiyuan University of Technology, 2007, 38(1): 35–37 [译自: 太原理工大学学报]     

一种基于信息最大化的自适应变步长盲源分离算法

现有的盲源分离算法不适合于数据的实时处理,并且算法性能依赖于步长的选择.提出一种基于信息最大化的自适应变步长盲源分离算法,采用基于估计函数的变步长算法,降低了盲源分离算法性能对步长的依赖性,并且采用自适应处理形式,适合数据的实时处理.最后将其应用于声音信号的盲分离,在选择小的步长参数的情况下,原有算法和文中新算法都取得了良好的分离效果;在选择较大的步长参数的情况下,新算法优于传统算法.     

基于改进l1范数最小化组合算法的欠定盲源分离

基于稀疏假设,欠定盲源分离问题一般可采用线性规划、最短路径法和组合算法等l1范数最小化方法进行求解,但是这些传统方法对源信号的稀疏性要求较高,从而限制了源信号的估计精度。为此,本文提出了一种改进的l1范数最小化组合算法.该算法根据一定阈值找到与最小l1范数解最接近的若干次优解,将这些次优解和最小l1范数解进行加权叠加,并替代最小l1范数解,作为源信号的估计。采用语音信号的仿真实验表明,对于观测信号个数不太小的高维混合情况,该算法的源信号估计精度能够比传统的l1范数最小化组合算法提高10%左右。     

基于FastICA算法的电力线通信OFDM信号分离方法

提出了一种基于快速独立成分分析(FastICA)算法的电力线通信正交频分复用(OFDM)技术信号分离方法。该方法在FastICA算法对2路观测信号进行解混处理的基础上,将解混信号的前导字段幅度归一化后分别与OFDM符号的前导序列的对应字段进行互相关,取相关值中最大值较大的解混信号作为通信信号,实现了通信信号和噪声的正确识别。定义分离比值为2路相关值中最大值的比值,作为解混效果的评价标准。通过设置合适的分离比值阈值,实现了期望的解混效果。以解混信号的前导字段为参考相位,采用DBPSK调制解调方式,实现了接收数据与解混信号相位无关。建立了MATLAB通信系统仿真模型,仿真结果验证了该方法在低信噪比条件且无纠错编码功能时仍具有较好的通信效果。最后基于硬件平台对算法的可实现性进行了验证测试。     

基于EMD盲源分离算法的电力系统低频振荡模式识别

针对广预测量系统低频振荡过程中的高斯噪声干扰和定阶问题,提出了基于EMD(empirical mode decomposition)盲源分离(blind source separation,BSS)算法的单通道低频振荡信号的模式分析方法。首先将信号利用经验模态分解得到一系列本征模函数分量组合的新信号;其次针对存在模态混叠的本征模函数分量,提出利用信号周期性构造其多路信号,并利用独立分量分析消除模态混叠的有效方法;然后利用盲源分离技术--二阶盲辨识算法(second order blind identification,SOBI),处理多通道观测信号矩阵,从中提取出不同的单模式信号;最后将去噪、定阶后的信号运用最小二乘-旋转不变技术(TLS-ESPRIT)算法辨识,得到低频振荡模态参数。数值算例仿真、IEEE四机两区域仿真实验表明该算法能够有效分离源信号,相比于其他方法具有抗噪性能好、拟合精度高等优点。     

Exploiting all combinations of microphone sensors in overdetermined frequency domain blind separation of speech signals

A new approach to overdetermined frequency domain blind source separation (BSS) of speech signals which exploits all combinations of observations and hence varying inter microphone spacings is proposed. The observations are divided into subgroups so that conventional frequency domain BSS algorithms can be used. By evaluating the separation performance obtained from each group on the basis of approximately measuring the independence of separated signals, the output of the group that has the best performance among all groups on a frequency‐by‐frequency basis is chosen as the overall output. The separated signals of the overall system are then obtained by transforming their frequency domain representations into the time domain. Simulation results based on speech signals confirm that the proposed approach has better performance based on the performance index (PI) as compared with a conventional scheme using only one microphone group and an existing overdetermined frequency domain BSS algorithm. Copyright © 2010 John Wiley & Sons, Ltd.     

盲源分离在振动机械故障诊断中的应用

采用了盲源分离的方法对机械振动信号进行了研究。首先简要地介绍了盲信号分离基本原理。基于最大负熵的原理,建立了用独立分量分析方法估计分离矩阵的FastICA的模型,并给出了实现步骤,编制了相应的程序。对实际振动传感器采集到的信号并进行盲分离实验,不同特征的信号可以被分离开来,分离出的信号极大地保留了源信号的信息特征。实验结果表明此方法是有效的,可作为振动机械故障诊断的信号预处理方法.     

脉冲噪声环境下机械故障信号的盲分离

针对振动传感器在采集故障信号时,在α稳定分布脉冲噪声的干扰下,使得传统机械故障信号时频盲源分离算法性能退化的问题,提出了一种基于分数低阶和S时频变换的盲源分离新方法。该方法先对传感器测试信号进行分数低阶子空间预白化,再计算低阶化信号的S变换时频分布,最后通过联合近似对角化恢复各个部分的故障源信号。通过计算机仿真实例分析表明,该算法能有效抑制脉冲噪声影响,避免了二阶矩或高阶矩无穷大的缺限,盲源分离效果较好,具有良好的鲁棒性。     

基于盲源分离和小波滤波的胎儿心电检测

利用胎儿心电信号与母亲心电信号及其他干扰信号之间的统计独立性、非相关性,采用盲源分离算法将胎儿心电信号从复杂的背景干扰中分离出来,但分离出来的胎儿心电图信号仍然受到噪声的干扰。本文提出联合小波滤波与盲源分离的胎儿心电检测方法,首先采用盲源分离处理多传感器采集到的信号,得到受干扰的胎儿心电图信号,然后采用小波滤除噪声,获得较为清晰的胎儿心电图信号。实际采集数据试验表明了该方法的可行性和优越性。     

改进的互信息最小化非线性盲源分离算法

提出了一种改进的互信息量最小化非线性盲源分离算法,改善了优化算法在串音误差方面大等的不足。该方法利用自然梯度优化算法来优化目标函数,避免了对矩阵的求逆计算,减少了计算时间。此外,在网络参数优化的过程中引入了扰动信号,提高了非线性盲源分离算法的寻优能力。实验结果表明改进的非线性盲源分离算法是有效的,而且相对传统的非线性盲源分离方法具有较小的误差。     

改进的FastICA在盲图像分离中的应用

盲源分离（BSS）是信号处理领域的一个热点问题。独立分量分析（ICA）是一种基于高阶统计量的信号分析方法,它可以找到隐含在数据中的独立分量,已广泛应用于信号处理领域。为了有效地对混合图像进行盲源分离,介绍了一种基于改进的快速固定点算法（FastlCA）,对经过随机线性混合后的模糊图像进行盲源分离。仿真结果显示,该算法可以很有效地对线性混合图像进行盲源分离。     

New self‐normalized blind source separation networks for instantaneous and convolutive mixtures

The main blind source separation networks proposed in this paper apply to convolutive mixtures (including instantaneous ones). They have a recurrent or direct structure and they may use channel‐specific separating functions. They are based on a self‐normalized weight adaptation rule, which adaptively estimates the average powers of non‐linear functions of the network outputs. This allows us to control several aspects of the operation of these networks, especially their convergence speed/accuracy trade‐off. It also makes them more robust with respect to non‐stationary situations. We analyse their convergence properties. We validate all these results by means of experimental tests performed with these networks, classical ones, and additionally proposed linear instantaneous direct networks based on a normalization of their outputs. These tests especially show that the proposed networks improve the convergence trade‐off and that only these networks apply to highly mixed non‐stationary sources. Copyright © 2002 John Wiley & Sons, Ltd.     

基于小波包混合优化的欠定盲源分离方法

为解决欠定盲源分离问题,提出一种基于小波包混合优化的欠定盲源分离方法。 该方法采用小波包变换将观测信号分 解,将观测信号的维数进行扩展,利用互相关系数值剔除冗余的信号分量,欠定盲源分离问题得到转化。 接着使用贝叶斯信息 准则下的奇异值分解方法估计源信号数目,通过白化过程对信号降维。 最后,引入鲸鱼优化算法中的螺旋泡网狩猎行为与莱维 飞行策略,对灰狼优化算法进行改进,将改进后的混合灰狼优化算法与独立成分分析算法相结合,实现重构正定白化信号的分 离,从而得到源信号的近似估计。 通过仿真实验对算法性能进行测试,结果验证所提方法的可行性和有效性。     

基于ICA盲源分离的声表面波无线传感器信号抗干扰方法

针对声表面波传感器无线信号易受环境中同频信号干扰的问题,设计了基于独立分量分析(ICA)盲源分离的抗干扰算法.该算法对传感器和同频干扰的混合信号进行分离,然后基于分离信号波形的衰减和等幅特征对信号进行判别.MATLAB仿真结果表明,算法能够有效分离混合信号并且保留了源信号的时域波形特征和频域信息.将抗干扰算法在基于数字...     

多用户DS/CDMA系统下的PN序列盲估计算法

本文提出了一种在非合作条件下多用户DS/CDMA系统的PN序列估计算法.该算法利用DS/CDMA系统独有的相关特性,在基带条件下对收集到的混合信号分段,利用不同分段数据码片两两相加的算法,实现多用户特征序列的快速分离.该方法不同于以往的神经网络算法,不需要用递归算法,从而简化了计算量,并且能对呈现高斯分布的伪随机序列进行有效的分离,并且该算法还有以下特性:不用对混合矩阵进行白化处理,可以保留了信道内多用户的幅值特征;该算法计算量几乎不受到用户扩频码位数影响,但相同误码率条件下较ICA算法需要稍高的信噪比;由于未采用递归,不存在收敛问题.计算机仿真结果表明,在信噪比SNR=0dB条件下,可以正确分离多用户特征码.     

基于一维盲源分离的滚动轴承故障诊断

盲源分离较之传统的信号处理方法在处理弱信号问题上更具优势。针对轴承故障诊断中因条件限制仅能进行单通道信号采集的情况,提出了一种基于总体经验模式分解的一维盲源分离算法。算法先通过总体经验模式分解将信号分解为多个本征模态函数,再根据本征模态函数之间的相关系数重组观测矩阵,最后利用近似联合对角化对矩阵进行盲源分离。通过数据仿真将该方法与小波分析和Hilbert-Huang变换作对比,说明该方法更适于处理低信噪比的轴承故障信号。对滚动轴承进行了故障诊断实验,成功找到了表征内圈故障和外圈故障的特征信息。     

断路器操动状态声音辨识的优化算法的研究

为解决基于声信号分析的高压断路器在线诊断故障中外界环境干扰问题,提出了一种声音扰动信号辨识的盲源分离方法。首先利用改进的势函数法进行源数估计,然后通过集合经验模态分解(EEMD)算法得到多个IMF分量,重构形成符合聚类源数的多维信号,并利用拟牛顿法优化快速独立分量分析算法,实现声音信号的盲源分离;最后根据包络特征比对获取断路器状态辨识的合闸声音信号分量。实验结果证明,文章提出的方法能够在信源信息未知情况下,从混叠声音信号有效地提取断路器操动产生的有用声信号。     

Blind separation of delayed instantaneous mixtures: a cross‐correlation based approach

A cross‐correlation based method is proposed for blind separation of statistically uncorrelated i.i.d. signals. In contrast to much of the existing work in the area, the proposed method allows the separation of more sources than sensors and the sensors are not restricted to have non‐Gaussian distributions. The mixed signals are modelled as a sum of attenuated and delayed source signals. As compared to the delayless mixture model commonly employed in the literature, incorporating delays in the model may offer better fit to practical applications, such as source separation in an anechoic environment. We estimate the delay and attenuation parameters from the peak locations and strengths of the cross‐correlation function, respectively. Using these parameters, we then discuss the use of four methods for the recovery of source signals. These methods are compared and their usage is proposed under different operating conditions. Copyright © 2004 John Wiley & Sons, Ltd.