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]     

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.     

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。     

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。     

多变量数据分析及应用研究

在统计信号处理及其相关领域,多变量数据的描述和分析一直是人们广泛关注的研究课题.在现有的多变量数据分析方法中,基于二阶统计特性的主分量分析(PCA)和基于高阶统计特性的独立分量分析(ICA)是两种非常有代表性的方法.本文在简要介绍PCA和ICA基本原理的基础上,结合脑电消噪问题,对两种方法的性能和特点进行了较深入地比较.实验结果表明,在非高斯信号处理上,独立分量分析方法具有明显的优势.     

基于形态滤波和独立分量分析的轴承故障盲分离

针对独立分量分析（ICA）对噪声较为敏感及滚动轴承故障信号的调制特性,提出一种基于形态滤波与ICA相结合的方法。该方法首先对观测信号进行形态滤波以突出故障特征同时消除其他干扰源,然后应用ICA分离形态滤波后信号。对滚动轴承外圈内圈复合故障信号进行实验研究,结果表明该方法能够有效识别分离滚动轴承故障特征。     

应用核独立分量分析的电力用户负荷曲线估计

提出了核独立分量分析算法,即白化的核主分量分析加上独立分量分析算法。该算法在电网信息不足时,利用电网部分支路的潮流作为观测值,就可以完成对用户负荷曲线的估计。经过IEEE 9节点系统的仿真验证,结果表明,观测值在经过白化的核主分量分析算法处理后,非高斯性增强。应用独立分量分析算法对处理后的观测值进行盲源分离后,所得用户负荷需求曲线的估计值逼近实际值。与仅用独立分量分析方法的仿真结果相比,估计误差降低,相关系数增加。     

用独立分量分析消除工频通信中的谐波干扰

通信信号在配电网传输中，其背景信号必然含有大量的谐波成分，这给通信信号的检测带来极大的困难。文中首先介绍了基于配电网的双向工频通信系统，并根据双向工频通信的信号特征，提出了基于独立分量分析的谐波消除方法。该方法在消除谐波干扰的同时，几乎对有用信号成分未有任何破坏影响。还通过MATLAB仿真对此方法进行了验证，结果良好。     

A new blind source separation method based on fractional lower‐order statistics

We proposed neural network structures related to multilayer feed‐forward networks for performing blind source separation (BSS) based on fractional lower‐order statistics. As alpha stable distribution process has no its second‐ or higher‐order statistics, we modified conventional BSS algorithms so that their capabilities are greatly improved under both Gaussian and lower‐order alpha stable distribution noise environments. We analysed the performances of the new algorithm, including the stability and convergence performance. The analysis is based on the assumption that the additive noise can be modelled as alpha stable process. The simulation experiments and analysis show that the proposed class of networks and algorithms is more robust than second‐order‐statistics‐based algorithm. Copyright © 2006 John Wiley & Sons, Ltd.     

Large‐scale super‐Gaussian sources separation using Fast‐ICA with rational nonlinearities

Independent component analysis (ICA) is one of the most powerful methods for solving blind source separation problem. In various ICA methods, the Fast‐ICA is an excellent algorithm, and it finds the demixing matrix that optimizes the nonlinear contrast function. There are three original contrast functions in the Fast‐ICA to separate super‐Gaussian and sub‐Gaussian sources, and their respective derivatives are similar to nonlinearities used in neural networks. For the separation of large‐scale super‐Gaussian sources, however, the contrast functions and the nonlinearities are not optimal owing to high computational cost. To solve this potential problem, this paper proposes four rational polynomial functions to replace the original nonlinearities. Because the rational polynomials can be quickly evaluated, when they are used in the Fast‐ICA, the computational load of the algorithms can be effectively reduced. The proposed rational functions are derived by the Pade approximant from Taylor series expansion of the original nonlinearities. To reduce the error of approximation, we make the behaviors of rational functions approach that of the original ones within an effective range as well as possible. The simulation results show that the Fast‐ICA algorithms with rational nonlinearities not only can speed up the convergence but also improve the separation performance of super‐Gaussian blind source separation. Copyright © 2016 John Wiley & Sons, Ltd.     

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.     

基于盲源分离技术的电力用户负荷曲线估计

盲源分离技术可以在不知道网络参数或拓扑结构的情况下,仅用部分支路有功潮流估计出各节点负荷.本文采用了核独立分量分析算法进行负荷曲线估计,使用再生核希尔伯特空间上的典型相关性分析作为目标函数.实验基于IEEE-14母线系统,估计出了8个负荷节点的有功功率.结果表明,同独立分量分析方法得到的仿真结果相比,核独立分量分析算法估计的结果误差更小,相关系数更大.     

Blind source separation algorithm for communication complex signals in communication reconnaissance

Most blind source separation algorithms are only applicable to real signals, while in communication reconnaissance processed signals are complex. To solve this problem, a blind source separation algorithm for communication complex signals is deduced, which is obtained by adopting the Kullback-Leibler divergence to measure the signals’ independence. On the other hand, the performance of natural gradient is better than that of stochastic gradient, thus the natural gradient of the cost function is used to optimize the algorithm. According to the conclusion that the signal’s mixing matrix after whitening is orthogonal, we deduce the iterative algorithm by constraining the separating matrix to an orthogonal matrix. Simulation results show that this algorithm can efficiently separate the source signals even in noise circumstances. __________ Translated from Journal of Huazhong University of Science and Technology (Natural Science Edition), 2007, 35(4): 33–36 [译自: 华中科技大学学报(自然科学版)]     

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。     

基于独立分量分析的谐波估计和消除

从盲源分离的角度对谐波估计和消除方法进行了探讨 ,提出了基于独立分量分析的谐波消除新方法。新方法的特点是在消除谐波干扰的同时 ,有效地保护有用信号成分不被破坏。文中通过实测数据对新方法进行了验证 ,并与传统陷波谐波消除方法进行了比较 ,取得了较理想的效果。     

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.     

基于独立分量分析的谐波检测

长期以来,谐波治理一直是电能质量控制的重要组成部分, 而准确的检测又是有效治理和分析谐波的前提和基础。将谐波的问题视作盲源分离问题,并将该领域中广泛使用的独立分量分析法运用到检测算法中。通过构建适当的虚拟观测源,从观测信号中分离出基波和各次谐波分量。实验结果表明,该方法在实时性要求不高的情况下可准确检测谐波。     

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.     

独立分量分析在谐波源辨识中的应用

为了确定电力系统谐波污染责任和治理责任,提出一种基于独立分量分析ICA(independent component analysis)的谐波源辨识方法.利用独立分量分析可以将相互独立的源信号从其线性混合的信号中分离出来的特点,将谐波源看作相互独立的源信号,对其混合后的电压信号进行计算可分离出相应的谐波源信号,完成谐波源的辨识.在Matlab软件中对电弧炉、饱和变压器这类典型的谐波源进行仿真并验证.结果表明该方法有很好的检测和判别效果.     

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。