基于信号稀疏表示的非参数基函数特征提取理论研究进展和展望

有效的信号特征提取技术在信号的分析、增强、压缩、复原等领域起着重要的作用,是模式识别、智能系统和故障诊断等诸多领域的基础和关键。虽然目前研究者提出了很多方法来解决这个问题,然而处理效果并不理想。非参数基函数特征提取是一种基于稀疏表示的特征提取方法,方法的核心是将观察信号分解为一组最好匹配信号局部结构的特征波形的线性展开,这些特征波形是由非参数基函数特征波形估计方法计算所得。详细描述了非参数基函数特征提取方法的理论思想,介绍了该方法的最新研究进展及其存在的问题,最后指出了该方法进一步发展的方向。     

非参数特征提取中模板信号的选取

非参数特征提取方法虽然不需要用任何参数表达的基函数,但在很大程度上却依赖于模板信号的选取。模板信号与实际信号逼近的程度,直接影响了提取结果的精度,这严重阻碍了该方法的广泛应用,因而模板信号的选取成为非参数波形提取中的一个关键问题。利用非参数特征提取方法前一次提取的结果,引入自适应调节模板信号的算法,使得该方法不再过多地要求模板信号具有信号的先验知识,提高了该方法在应用中的柔性和适应性。仿真信号表明了所提方法的可行性和有效性。研究结果为非参数特征提取方法的应用提供了一条新途径。     

SNR scalability based on matching pursuits

In this paper, SNR scalable representations of video signals are studied. The investigated codecs are well suited for communications applications because they are all based on backward motion-compensated predictive coding, which provides the necessary low-delay property. In a very-low bit rate context (VLBR), the matching pursuits (MP) signal representation algorithm is used to represent the displaced frame difference (DFD) of each layer of a multilevel decomposition of the video signal. A number of conventional prediction schemes that can be generalized to any DFD representation technique are considered. They are compared with an original and MP specific DFD prediction method. Two scenari have been considered. In the first scenario, an enhancement layer is built on a base layer that has been encoded using a classical, i.e., nonscalable scheme. In that case, all methods appear to be comparable, In the second scenario, the fact that the base layer is used as a reference for an enhancement layer is taken into account to build it. In that case, the proposed MP prediction method clearly outperforms all other conventional approaches, Additional lessons can be drawn from this work. The same motion vectors can be used in both SNR layers, and the DFD prediction between layers improves coding efficiency. Moreover, the MP representation of the signal enable us to measure the predictability of the high SNR layer DFD from the low SNR layer DFD, i.e., to quantify the part of the low SNR layer information that also belongs to the high SNR layer     

Clustering of electrocardiograph signals in computer-aided Holter analysis

A number of methods for temporal alignment, feature extraction and clustering of electrocardiographic signals are proposed. The ultimate aim of the paper is to find a method to automatically reduce the quantity of beats to examine in a long-term electrocardiographic signal, known as Holter signal, without loss of valuable information for the diagnosis. These signals include thousands of beats and therefore visual inspection is difficult and cumbersome. All the elements involved in each stage will be described and a thorough experimental study will be presented. The electrocardiograph signals used in the experiments belong to the well-known MIT database, where many different waveforms can be found. Based on the results of the experiments, a complete process is proposed to obtain the significant beats present within a signal, with a reasonable computational cost. Hence, cardiologists will only have to examine a small but fully representative subset of beats, making this method a very useful tool for medical decision support systems.     

A robust extraction algorithm for biomedical signals from noisy mixtures

Blind source extraction (BSE) is widely used to solve signal mixture problems where there are only a few desired signals. To improve signal extraction performance and expand its application, we develop an adaptive BSE algorithm with an additive noise model. We first present an improved normalized kurtosis as an objective function, which caters for the effect of noise. By combining the objective function and Lagrange multiplier method, we further propose a robust algorithm that can extract the desired signal as the first output signal. Simulations on both synthetic and real biomedical signals demonstrate that such combination improves the extraction performance and has better robustness to the estimation error of normalized kurtosis value in the presence of noise.     

Noisy component extraction with reference

Blind source extraction (BSE) is particularly attractive to solve blind signal mixture problems where only a few source signals are desired. Many existing BSE methods do not take into account the existence of noise and can only work well in noise-free environments. In practice, the desired signal is often contaminated by additional noise. Therefore, we try to tackle the problem of noisy component extraction. The reference signal carries enough prior information to distinguish the desired signal from signal mixtures. According to the useful properties of Gaussian moments, we incorporate the reference signal into a negentropy objective function so as to guide the extraction process and develop an improved BSE method. Extensive computer simulations demonstrate its validity in the process of revealing the underlying desired signal.     

基于小波系数聚类的特征提取分类方法

神经网络是一种普遍采用的模式分类方法,当对样本的抽样数目较大时,神经网络结构复杂,训练时间激增,分类性能下降,针对这一问题,提出一种基于快速小波变换特征提取的分类方法。着先对婆婆以系数矩阵的每行进行聚类,表达重要频率范围内小波系数矩阵的行有较多的聚类数,从而大大减少了神经网络的输入数,而同时保留了有用的信息。特征提取后,采用小波系数的能量值特征量,应用径向基函数网络识别肺发出的各种不同的声音,实验证明：该方法有较高的识别率。     

适合盲提取的自适应遗传算法

对盲分离问题中存在收敛速度慢、精度不高和容易陷入局部最优等缺点进行了研究,提出了一种基于改进自适应遗传算法的快速盲提取算法。在负熵判据的基础上,建立了最小化独立信号边缘熵准则。以盲提取目标优化函数为基础,对遗传算法的关键技术进行了改进,同时提出一种适合盲信号提取的适应度函数和防止算法局部收敛的监测策略,使算法能够自动跳出局部最优,快速地收敛于全局最优解。以改进的自适应遗传算法作为寻优算法,快速地实现了瞬时混合信号的盲提取。仿真实验表明,该算法性能稳定、收敛速度快,得到了全局最优解,有效地实现了信号盲提取。     

Classification of modulation signals using statistical signal characterization and artificial neural networks

Modulation recognition systems have to be able to correctly classify the incoming signal modulation scheme in the presence of noise. A new method for classification of analogue and digital modulated signals at low signal-to-noise ratio (SNR) is introduced in this paper. This method uses the statistical signal characterization (SSC) to extract parameters to classify the different modulation signals. The SSC technique produces a set of four numerical parameters for a specific modulated signal. Subsequent comparison of these parameters to those of other waveforms provides the basis for our classification system. The results of SSC technique are applied to an artificial neural network (ANN) to have a robust classification system in the presence of noise down to SNR of 3 dB. No a priori information is required by this technique about the set of input waveforms. The input to the classification system can be analogue or digital signals or a combination of both. The proposed technique shows a 100% efficiency of classification of analogue signals or digital signals at SNR of 7 dB. This classification efficiency reduces to 83% and 86% for analogue or digital signals at SNR of 3 dB. The SSC technique shows better classification results in comparison with other techniques with an important advantage over other methods, which is the simplicity of the neural network needed with this technique due to the small number of features used in the classification.     

基于高斯平滑与模糊函数等高线的雷达辐射源信号分选

雷达辐射源信号分选是电子侦察系统、威胁告警系统的关键步骤.针对现有基于模糊函数的复杂体制雷达辐射源信号分选方法信息利用率低、易受噪声影响等问题, 提出一种基于模糊函数等高线的分选新方法; 首先, 对信号的模糊函数进行高斯平滑处理并绘制其等高线作为进一步的特征提取对象; 其次, 从图像处理的角度提取正外接矩和方向角作为雷达信号分选的特征向量; 最后, 用核模糊C均值聚类算法对特征向量进行分选.仿真实验表明, 所提方法在8 dB以上的固定信噪比环境下分选6类典型信号的成功率均为100 %, 即使在0 dB环境下, 分选成功率也保持在89.04 %以上; 在0 ~ 20 dB动态信噪比环境下分选成功率达到96.36 %.实测数据验证, 所提特征提高了5种外场辐射源信号的分选效果, 可作为经典5参数的有效补充. 此外, 所提特征还具备较低的计算量, 提取单个信号特征的耗时仅为0.24 s, 具有一定的工程价值.     

基于K-SVD超声渡越时间获取方法研究

在超声回波检测信号中,反映污垢特征的冲击信号非常微弱,容易被噪声淹没。针对信号稀疏分解中常用匹配追踪分解不够准确的问题,提出基于K-SVD奇异值分解的超声渡越时间获取方法,利用K-SVD训练得到超声回波信号的过完备字典,结合正交匹配追踪进行局部搜索适配原子,以提高信号稀疏分解的速度和准确度。基于Comsol Multipysics仿真软件建立充液污垢管道三维有限元模型,研究了超声回波传播特性规律。将K-SVD算法应用于超声回波仿真信号和换热污垢管道回波检测信号的处理,并与原始小波训练字典进行对比。结果表明：改进的K-SVD字典学习算法能够在提高信号稀疏分解的同时,获得较好的降噪结果和污垢特征信息提取,对超声检测信号的处理具有实际意义。     

A new optimum feature extraction and classification method for speaker recognition: GWPNN

Speech and speaker recognition is an important topic to be performed by a computer system. In this paper, an expert speaker recognition system based on optimum wavelet packet entropy is proposed for speaker recognition by using real speech/voice signal. This study contains both the combination of the new feature extraction and classification approach by using optimum wavelet packet entropy parameter values. These optimum wavelet packet entropy values are obtained from measured real English language speech/voice signal waveforms using speech experimental set. A genetic-wavelet packet-neural network (GWPNN) model is developed in this study. GWPNN includes three layers which are genetic algorithm, wavelet packet and multi-layer perception. The genetic algorithm layer of GWPNN is used for selecting the feature extraction method and obtaining the optimum wavelet entropy parameter values. In this study, one of the four different feature extraction methods is selected by using genetic algorithm. Alternative feature extraction methods are wavelet packet decomposition, wavelet packet decomposition – short-time Fourier transform, wavelet packet decomposition – Born–Jordan time–frequency representation, wavelet packet decomposition – Choi–Williams time–frequency representation. The wavelet packet layer is used for optimum feature extraction in the time–frequency domain and is composed of wavelet packet decomposition and wavelet packet entropies. The multi-layer perceptron of GWPNN, which is a feed-forward neural network, is used for evaluating the fitness function of the genetic algorithm and for classification speakers. The performance of the developed system has been evaluated by using noisy English speech/voice signals. The test results showed that this system was effective in detecting real speech signals. The correct classification rate was about 85% for speaker classification.     

Waveform detection with RBF network – Application to automated EEG analysis

Automated detection of different waveforms in physiological signals has been one of the most intensively studied applications of signal processing in the clinical medicine. During recent years an increasing amount of neural network based methods have been proposed. In this paper we present a radial basis function (RBF) network based method for automated detection of different interference waveforms in epileptic EEG. This kind of artefact detector is especially useful as a preprocessing system in combination with different kinds of automated EEG analyzers to improve their general applicability. The results show that our neural network based classifier successfully detects artefacts at the rate of over 75% while the correct classification rate for normal segments is as high as about 95%.     

Audio watermarking scheme based on embedding strategy in low frequency components with a binary image

With the recent development of information technology and computer network, digital format of data has become more and more popular. However, a major problem faced by digital data providers and owners is protecting data from unauthorized copying and distribution. As a solution to the problem, digital watermark technology is now attracting attention as new method of protection against said unauthorized copying and distribution. The aim of the digital audio watermarking is to take prespecified data that carries certain information and hide it within the audio stream such that it is not audible to the human ear (i.e., transparent) but at the same time renders the file more resistant to removal (i.e., robust). In this paper, we propose a new method for embedding digital watermarks into audio signals in low frequency components, which method mitigates these and other related shortcomings. The proposed method uses the wavelet transform constructed by lifting-based wavelet transform (LBWT) in order to provide a fast implementation between watermark embedding and extraction parts. In the first stage of the proposed method, the original audio host signal is converted to a wavelet domain using LBWT. The signal is thus decomposed into low and high frequency components. Approximation coefficients correspond to low frequency components of the signal. Next, the watermark generated by pseudorandom numbers is embedded into wavelet approximation coefficients of the segmented host audio signal depending on the binary value of the binary image. The reason for embedding the watermark in the low frequency components is that these components' energy is greater than that of high frequency components in such a way that the watermark is inaudible; therefore, it should not alter the audible content and should not be easy to remove. The proposed method uses a binary image to decide whether or not the watermark generated by pseudorandom numbers is embedded in the audio host signal. To evaluate the performance of the proposed audio watermarking method, subjective and objective quality tests including bit error rate (BER) and signal-to-noise ratio (SNR) are conducted. The tests' results show that the proposed method yields a high recovery rate after attacks by commonly used audio data manipulations such as low-pass filtering, requantization, resampling and MP3 compression.     

Constant false alarm rate detection of saccadic eye movements in electro-oculography

The analysis of eye movements has proven to be valuable in both clinical work and research as well as in other fields besides medicine. The detection of saccadic eye movements and the extraction of related saccade parameters, such as maximum angular velocity, amplitude, and duration, are usually performed during the analysis of electro-oculographic (EOG) signals. This article considers a saccade detection method that is based on the constant false alarm rate technique, in which the detection sensitivity is continuously adjusted on the basis of the observed signal in order to keep the number of false alarms constant. The method is computationally efficient, it can operate autonomously without user intervention, and it is capable of detecting saccades in a sequential fashion. Therefore, the method finds potential use in applications that require automated analysis of electro-oculographic signals. Because of the constant false alarm rate property, the method can also perform in situations where ideal measurement conditions cannot be guaranteed and noise presents a considerable problem.     

Onset detection of ultrasonic signals for the testing of concrete foundation piles by coupled continuous wavelet transform and machine learning algorithms

The construction of ultra-high-rise and long-span structures requires higher requirements for the integrity detection of piles. The acoustic signal detection has been verified an efficient and accurate nondestructive testing method. In fact, the integrity of piles is closely related to the onset time of signals. The accuracy of onset time directly affects the integrity evaluation of a pile. To achieve high-precision onset detection, continuous wavelet transform (CWT) preprocessing and machine learning algorithms were integrated into the software of high-sampling rate testing equipment. The distortion of waveforms, which could interfere with the accuracy of detection, was eliminated by CWT preprocessing. To make full use of the collected waveform data, three types of machine learning algorithms were used for classifying whether the data points are ambient or ultrasonic signals. The models involve a commonly used classifier (ELM), an individual classification tree model (DTC), an ensemble tree model (RFC) and a deep learning model (DBN). The classification accuracy of the ambient and ultrasonic signals of these models was compared by 5-fold validation. Results indicate that RFC performance is better than DBN and DTC after training. It is more suitable for the classification of points in waveforms. Then, a detection method of onset time based on classification results was therefore proposed to minimize the interference of classification errors on detection. In addition to the three data mining methods, the autocorrelation function method was selected as the control method to compare the proposed data mining based methods with the traditional one. The accuracy and error analysis of 300 waveforms proved the feasibility and stability of the proposed method. The RFC-based detection method is recommended because of the highest accuracy, lowest errors, and the most favorable error distribution among four onset detection methods. Successful applications demonstrate that it could provide a new way for ensuring the accurate testing of pile foundation integrity.     

基于多变量符号转移熵的癫痫脑电分析

大脑神经元细胞群的异常同步放电是癫痫的病因,这种异常放电是目前诊断癫痫的重要依据。利用复杂度理 论来分析癫痫信号已经成为研究热点,而符号转移熵是反应系统混乱程度的一种非线性指标,在研究癫痫脑电信号特征的提取中有重要的作用。符号转移熵一般都是用来衡量两 个变量之间的动力学特征及方向性信息,忽略了多个变量之间相互作用。本文基于多变量符号转移熵研究分析了癫痫脑电信号,实验中将原始信号符号化后通过数值分析,对导联信号及信号长度的选取以及稳健性分析,表明该方法能够对正常人与癫痫病人的脑电信号进行显著区分,且该算法稳健可靠,该研究结果对临床辅助诊断有帮助。     

基于EMD和互信息熵的微震信号自适应去噪

针对强干扰背景下的微震信号提取,提出一种基于经验模态分解（Empirical Mode Decomposition,EMD）和互信息熵的自适应提取算法。通过EMD对微震信号进行分解,得到高频和低频两部分信号,并对分解得到的各阶固有模态分量求出能量和能量熵值。根据互信息准则,通过依次计算相邻分量能量熵之间的互信息值来区分高频和低频信号。将经过自适应阈值滤波后的高频信号和低频信号一起进行信号重构,得到新的微震信号。仿真结果表明,在对微震信号去噪时,该方法可以有效地去除噪声信号,信噪比均提升了10 dB以上。工程上的微震信号通过该方法处理后,也取得了较好的效果。     

基于多源Mallat-NIN-CNN网络的变压器绕组故障诊断

变压器绕组形变是常见的故障，传统的诊断方法参数过多且受噪音干扰导致诊断性能较差。提出了一种基于灰度转换的特征提取方法，该方法将振动信号转换为灰度图像，有效地提取特征。针对单源信号特性信息强度随距离变化的问题，利用多源通道采集振动信息，并利用图像融合手段抑制多源图像中大量冗余信息、信噪比低的问题，提出基于多源Mallat-NIN-CNN网络的电力变压器绕组故障诊断模型，利用Mallat算法对多源振动信号灰度图像分解，通过基于区域特性量测和加权平均方法分别对各分解层的高频分量和低频分量进行融合，将重构的灰度图像输入NIN-CNN网络进行故障诊断。经实验验证，该方法有效抑制了多源信号中的噪声，提高特征信息的完整性，降低了计算量，提高了故障诊断准确性。     

High-resolution time–frequency representation of EEG data using multi-scale wavelets

An efficient time-varying autoregressive (TVAR) modelling scheme that expands the time-varying parameters onto the multi-scale wavelet basis functions is presented for modelling nonstationary signals and with applications to time–frequency analysis (TFA) of electroencephalogram (EEG) signals. In the new parametric modelling framework, the time-dependent parameters of the TVAR model are locally represented by using a novel multi-scale wavelet decomposition scheme, which can allow the capability to capture the smooth trends as well as track the abrupt changes of time-varying parameters simultaneously. A forward orthogonal least square (FOLS) algorithm aided by mutual information criteria are then applied for sparse model term selection and parameter estimation. Two simulation examples illustrate that the performance of the proposed multi-scale wavelet basis functions outperforms the only single-scale wavelet basis functions or Kalman filter algorithm for many nonstationary processes. Furthermore, an application of the proposed method to a real EEG signal demonstrates the new approach can provide highly time-dependent spectral resolution capability.