小波盲源分离算法的仿真及FPGA实现

提出了一种基于小波变换的盲源分离方法,在理论分析和仿真结果的基础上,给出了FPGA的实现方案.针对传统盲分离算法对源信号统计特征敏感的弱点,该方法在小波变换域实施盲分离算法.同时,在分离过程中引入了空域相关滤波,使本文方法具有抗噪特征.将该方案应用于多接收天线系统中的多跳频信号的分离应用中,仿真结果证明了该方案的可行性和有效性.     

利用遗传粒子滤波的单通道扰信盲分离算法

针对低信噪比条件下单通道通信信号与干扰盲分离问题,本文提出了一种基于遗传粒子滤波的单通道盲扰信分离新算法.该算法首先建立了受扰信号的状态空间模型,并利用粒子滤波得到通信码元和未知参数的最大后验估计.针对标准粒子滤波中存在的粒子退化现象,本文引入了遗传进化操作来迭代估计优质粒子,在减少了所需粒子数量的同时,又保持了序贯估计过程中粒子集合的多样性和优质性,使新算法在低信噪比条件下具有更好的分离效果.仿真结果表明,新算法在干信比小于15 dB,信噪比大于10 dB的条件下,可以有效地从单路接收的受扰信号中分离出通信信号与干扰.     

肌电三维仿真手的单通道盲识别系统

介绍的肌电三维仿真手的单通道盲识别系统.利用单通道盲源分离,避免硬件电路连线复杂;利用混沌加密,确定表面肌电信号(Electromyography)盲源识别顺序.实验表明,该系统实现了肌电信号单通道纪录,解决了盲源分离中的分离无序的问题,准确地实现手部运动模式的实时识别和显示.     

基于独立成分分析的跳频信号盲分离

近年来,ICA（Independent Component Analysis,独立成分分析）已成为处理BSS（Blind Source Separation,盲源分离）问题的主要手段,同时也受到人们越来越多的关注。该文首先介绍ICA,然后引入FastICA算法的推导过程,最后通过MATLAB仿真将跳频信号进行盲分离,并与梯度算法所得的仿真结果进行对比分析。通过算法验证,经FastICA处理得到的分离信号与源信号相关系数的绝对值不小于0.99,与梯度算法比较可以明显地得到FastICA是一种更为有效的跳频信号盲分离方法。     

基于欠定盲分离的同步/异步跳频网台分选方法

研究无线通信网台分选优化问题,现有的跳频网台分选大多要求阵元数必须大于信号数,由于在实际应用中常常很难满足,为了用有限的阵元数分选尽可能多的跳频信号,提出了一种基于欠定盲分离的跳频信号分选方法.根据跳频信号具有良好的时频稀疏性,依据稀疏分量分析的“两步法”思路,首先用单源点聚类方法估计混合矩阵;然后利用子空间投影方法分离各跳频信号,从而实现了跳频信号的欠定盲分离.仿真结果表明了改进方法的有效性,为通信网分选优化提供了参考.     

基于经验模态分解的单通道盲源分离算法

为了提高单通道盲源分离性能,首先由单路信号利用经验模态分解得到一系列本征模函数分量组合成多路信号;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;然后利用互相关性消除上述所得到的多路信号中的虚假分量,并将剩余的分量信号与观测信号构成新的多路信号;最后利用Fast-ICA(fast-independent component analysis)算法分离得到源信号。仿真实验表明该算法能够有效分离源信号,分离性能优于目前已有的基于经验模态分解的单通道盲源分离算法。     

基于时频分布的非平稳信号盲分离算法

针对传统的盲源分离方法往往忽略信号非平稳性的问题,基于从瞬时线性混合模型的观测信号中分离出相互独立的源信号,并针对信号具有非平稳性,结合时频分析和盲源分离各自的特点,对非平稳信号盲分离进行了研究,并提出了一种新的具有不同空间时频分布的非平稳盲分离算法.仿真实验表明,通过采用维纳全时频域搜索来寻找局部最大值的平滑伪Wigner-Ville分布,该算法可以抑制交叉项而且能够保持时频聚集性,并达到了很好的分离效果.     

一种基于一维卷积残差网的单通道信号源个数估计算法

单通道信号源个数估计是单通道盲源分离问题的前提与难点,传统方法无法直接进行估计且准确率较低.文章提出了一种基于深度网络分类器的单通道信号源估计方法.该方法将源个数估计作为分类问题,在经典CNN的基础上引入一维卷积网络与残差结构作为分类器,采用短时傅里叶变换和梅尔倒谱系数作为联合特征输入分类器.在Libricount数据集上的测试结果表明,该方法的源个数估计准确率明显优于基准模型.     

基于EEMD和ICA的单通道列车信号盲分离

针对列车混合故障的诊断,提出了一种基于集合平均经验分解（EEMD）和独立分量分析（ICA）的盲分离诊断方法。通过EEMD算法将混合信号分解为包含不同源信号特征的本征模态函数（IMF）,组成新的多维信号;用主成分分析准确估计源信号个数,解决了单通道信号盲分离的欠定问题;利用快速独立分量分析（Fast-ICA）算法实现了信号的盲分离。实验信号分别采用仿真信号和列车实验信号进行实验,实验结果表明,该算法可以有效地分离出列车的单故障信号。     

语音增强的计算机仿真

研究关于盲源分离的特征向量分离算法在语音增强的应用,传统的方法对混合的语音信号很难进行有效的分离,而在实际中很多场合都需要对语音信号进行增强.为消除噪音,提高清晰度,使用的盲源分离算法却正能实现传统方法难以实现的技术.运用一种盲源分离的特征向量分离算法来进行语音增强,并且对实际的两个语音信号运用该算法进行了混合语音信号的分离增强实验,利用MATLLAB软件对混合语音信号进行了盲源分离的特征向量分离算法的仿真,可从混合语音信号分离出了两个原始语音信号.证明了盲源分离算法应用于语音分离的可行性,为盲源分离应用于语音增强提供了参考依据.     

An Efficient Measure of Signal Temporal Predictability for Blind Source Separation

An efficient measure of signal temporal predictability is proposed, which is referred to as difference measure. We can prove that the difference measure of any signal mixture is between the maximal and minimal difference measure of the source signals. Previous blind source separation (BSS) problem is changed to a generalized eigenproblem by using Stone’s measure. However, by using difference measure, the BSS problem is furthermore simplified to a standard symmetric eigenproblem. And the separation matrix is the eigenvector matrix, which can be solved by using principal component analysis (PCA) method. Based on difference measure, a few efficient algorithms have been proposed, which are either in batch mode or in on-line mode. Simulations show that difference measure is competitive with Stone’s measure. Especially, the on-line algorithms derived from difference measure have better performance than those derived from Stone’s measure.     

A BSS-based space–time multi-channel algorithm for complex-jamming suppression

This paper considers the problem of suppressing complex-jamming, which contains sidelobe blanket jammings (SLJs), multiple near-mainlobe blanket jammings (multiple-NMLJs) and self-defensive false target jamming (SDJ). We propose a blind source separation (BSS)-based space–time multi-channel algorithm for complex-jamming suppression. The space–time multi-channel consists of spatial multiple beams and temporal multiple adjacent pulse repetition intervals (PRIs). The source signals can be separated by the BSS, owing to their statistical independence. The real target and SDJ can then be obtained by the pulse compression approach, distinguished by echo identification simultaneously. A remarkable feature of the proposed approach is that it does not require prior knowledge about real target or jammings, and it is easy to implement for engineering applications.     

基于盲源分离的超声信号去噪的仿真研究

提出一种新的基于盲源分离的超声信号去噪方法.为了验证去噪方法的有效性,应用此方法处理了仿真的超声信号,并与小波去噪的效果进行了比较.实验结果表明:该去噪方法能极大提高超声信号的信噪比,且其效果能与小波去噪方法相媲美,其特点是通过超声信号和噪声信号的盲源分离实现噪声消除.     

盲分离与时频分析融合的多源信道均衡

针对多信源条件下强多途干扰严重的水声信道,提出一种盲分离与时频分析融合的多源信道均衡技术。该方法首先应用盲分离技术将多源接收信号进行有效分离,对分离后的信号进行时频分析运算,然后采用Radon变换加解线调方法估计出信号的主要参数,通过重构声源信号最终完成信号的复原。通过对计算机仿真数据和海试数据的处理,验证了该方法的可行性和有效性。     

Atrial fibrillatory signal estimation using blind source extraction algorithm based on high-order statistics

The analysis and the characterization of atrial fibrillation （AF） requires, in a previous key step, the extraction of the atrial activity （AA） free from 12-lead electrocardiogram （ECG）. This contribution proposes a novel non-invasive approach for the AA estimation in AF episodes. The method is based on blind source extraction （BSE） using high order statistics （HOS）. The validity and performance of this algorithm are confirmed by extensive computer simulations and experiments on realworld data. In contrast to blind source separation （BSS） methods, BSE only extract one desired signal, and it is easy for the machine to judge whether the extracted signal is AA source by calculating its spectrum concentration, while it is hard for the machine using BSS method to judge which one of the separated twelve signals is AA source. Therefore, the proposed method is expected to have great potential in clinical monitoring.     

基于Givens矩阵和联合非线性不相关的盲源分离新算法

针对源信号统计独立的盲源分离(Blind Source Separation,BSS)问题,提出了一种基于Givens矩阵和联合非线性不相关的盲源分离新算法.由于分离信号独立性的度量是影响算法有效性的重要因素,因此首先提出了一种改进的度量独立性的方法,该方法以独立源信号的联合非线性不相关来度量独立性;其次,结合Givens矩阵可以对分离矩阵施加正交性约束且能减少要估计参数个数的性质,将盲源分离问题转化成无约束优化问题,并利用拟牛顿法中的BFGS算法求解该无约束优化问题,得到分离矩阵;最后,通过模拟混合信号和真实语音混合信号的分离实验验证了该算法的有效性.     

Exploiting cyclostationary difference in anti-jamming method based on single channel source separation

Traditional time/frequency domain filter-based methods would fail due to the serious overlapping of the communication signal and jamming in time/frequency domain. In order to solve this problem, an anti-jamming algorithm based on single channel source separation by exploiting cyclostationary difference is proposed for direct sequence spread spectrum (DSSS) communication. Firstly, the cyclostationary difference among source signals is used to establish the single channel source separation model based on partial priori information of the DSSS signal. Secondly, cost function of the algorithm is constructed by the training sequences and the minimum mean square error criterion; after that, the optimal separation matrix of the pilot mixed signal can be calculated based on this cost function. Thirdly, the purpose of anti-jamming for DSSS communication can be achieved by separating the communication signal from the service mixed signal based on the optimal separation matrix calculated before. Simulation results show that, compared with the unseparated signal of high bit error rate (BER) close to 0.5, the proposed antijamming algorithm is more effective in anti-jamming with the BER as low as 10^–3, which is not only applicable to narrowband jamming such as the single-tone/multi-tone jamming, but also applicable to broadband jamming such as the sweep-spot jamming.     

基于递归神经网络结构的非平稳信号自适应盲分离

基于递归网络分离结构并利用时间相关的评价函数，针对二输入二输出盲信号分离问题，提出了一种非平稳信号的自适应盲分离算法。该算法计算量小，可根据输出信号能量大小有选择地更新分离系数。并可扩展到多输入多输出盲分离问题。仿真验证对声音等非平稳信号具有良好的分离效果。     

Blind source separation with unknown and dynamically changing number of source signals

The contrast function remains to be an open problem in blind source separation (BSS) when the number of source signals is unknown and/or dynamically changed. The paper studies this problem and proves that the mutual information is still the contrast function for BSS if the mixing matrix is of full column rank. The mutual information reaches its minimum at the separation points, where the random outputs of the BSS system are the scaled and permuted source signals, while the others are zero outputs. Using the property that the transpose of the mixing matrix and a matrix composed by m observed signals have the indentical null space with probability one, a practical method, which can detect the unknown number of source signals n, ulteriorly traces the dynamical change of the sources number with a few of data, is proposed. The effectiveness of the proposed theorey and the developed novel algorithm is verified by adaptive BSS simulations with unknown and dynamically changing number of source signals.     

非正交联合对角化的罚函数粒子群算法

针对多个矩阵近似联合对角化盲分离问题,提出一种新的非正交近似联合对角化算法.首先采用罚函数法将联合对角化的非线性约束优化模型转化为无约束优化模型;其次将粒子群优化算法引入无约束优化模型中实现目标函数的最优化,从而完成矩阵组的联合对角化.分析了惩罚因子的更新策略及算法的收敛性能,并设计仿真实验进行对比分析以检验算法解决实际盲分离问题的能力.