Blind Separation of Speech Signals Based on Wavelet Transform and Independent Component Analysis

Speech signals in frequency domain were separated based on discrete wavelet transform (DWT) and independent component analysis (ICA). First, mixed speech signals were decomposed into different frequency domains by DWT and the subbands of speech signals were separated using ICA in each wavelet domain; then, the permutation and scaling problems of frequency domain blind source separation (BSS) were solved by utilizing the correlation between adjacent bins in speech signals; at last, source signals were reconstructed from single branches. Experiments were carried out with 2 sources and 6 microphones using speech signals at sampling rate of 40 kHz. The microphones were aligned with 2 sources in front of them, on the left and right. The separation of one male and one female speeches lasted 2.5 s. It is proved that the new method is better than single ICA method and the signal to noise ratio is improved by 1 dB approximately.

Separation of cobalt and nickel by non-equilibrium solvent extraction

The separation of cobalt and nickel in the ammoniacal sulfate solution by non equilibrium solvent extraction with a phosphate (P303) as extractant was studied. In the experiment, the effects of equilibrium pH value in aqueous phase, contact time of the two phases, the air blowing time for feed liquor in the open beaker on percentage extraction of cobalt and nickel and percentage reextraction of nickel from the loaded organic phase with dilute H 2SO 4. etc were studied. The results showed that: Co(Ⅱ) can be oxidized to Co(Ⅲ) ammino complex by adding (NH 4) 2S 2O 8 or blowing air to the aqueous phase, and Co(Ⅲ) ammino complex is a kind of kinetically inert complex. Its extractive speed is very slow, while the nickel′s is much faster than that of cobalt. By controlling the contact time of the two phases, nickel can be separated from cobalt by non equilibrium solvent extraction. Then nickel was reextracted from the loaded organic phase with dilute H 2SO 4.

Blind source separation of multichannel electroencephalogram based on wavelet transform and ICA

Combination of the wavelet transform and independent component analysis （ICA） was employed for blind source separation （BSS） of multichannel electroencephalogram （EEG）. After denoising the original signals by discrete wavelet transform, high frequency components of some noises and artifacts were removed from the original signals. The denoised signals were reconstructed again for the purpose of ICA, such that the drawback that ICA cannot distinguish noises from source signals can be overcome effectively. The practical processing results showed that this method is an effective way to BSS of multichannel EEG. The method is actually a combination of wavelet transform with adaptive neural network, so it is also useful for BBS of other complex signals.

New Wavelet Threshold Denoising Method in Noisy Blind Source Separation

In general conditions, most blind source separation algorithms are established on noisy-free model and ignore the noise that affects the quality of separated sources. Firstly, this paper introduces an improved natural gradient algorithm based on bias removal technology to estimate the demixing matrix under noisy environment. Then the discrete wavelet transform technology is applied to the separated signals to further remove noise. In order to improve the separation effect, this paper analyzes the deficiency of hard threshold and soft threshold, and proposes a new wavelet threshold function based on the wavelet decomposition and reconfiguration. The simulations have verified that this method improves the signal noise ratio （SNR） of the separation results and the separation precision.

Study on the separation process of non-metallic inclusions at the steel-slag interface using water modeling

The separation process of non-metallic inclusions at the steel–slag interface was simulated by physical modeling. Three different kinds of particles (octahedral, plate-like, and spherical) and three different oils (kerosene, bean oil, and pump oil) were used to model inclusions and slags, respectively. The effects of inclusion geometry (shape and size) and slag properties (viscosity and interfacial tension) on the separation process were investigated. The results revealed that the variation of surface free energy and the viscosity of the slag are two significant factors affecting the separation process of inclusions at the steel–slag interface. The variation of surface free energy helped inclusions enter the slag phase, whereas the decrease of slag viscosity shortened the separation time. The deformation of the steel–slag interface could give rise to the resistance force, which would resist inclusions passing through the interface. A liquid film formed on the inclusion as it passed through the steel–slag interface, which might be related to the inclusion’s shape. © 2017, University of Science and Technology Beijing and Springer-Verlag Berlin Heidelberg.

EXISTENCE AND UNIQUENESS OF SOLUTIONS FOR NONLINEAR FRACTIONAL DIFFERENTIAL EQUATIONS WITH NON-SEPARATED TYPE INTEGRAL BOUNDARY CONDITIONS

In this paper, we study a boundary value problem of nonlinear fractional dif- ferential equations of order q （1 〈 q 〈 2） with non-separated integral boundary conditions. Some new existence and uniqueness results are obtained by using some standard fixed point theorems and Leray-Schauder degree theory. Some illustrative examples are also presented. We extend previous results even in the integer case q = 2.

COMPACTLY SUPPORTED NON-TENSOR PRODUCT FORM TWO-DIMENSION WAVELET FINITE ELEMENT

Some theorems of compactly supported non-tensor product form two-dimension Daubechies wavelet were analysed carefully. Compactly supported non-tensor product form two-dimension wavelet was constructed, then non-tensor product form two dimension wavelet finite element was used to solve the deflection problem of elastic thin plate. The error order was researched. A numerical example was given at last.

Ground roll separation method via threshold fi ltering and constraint of seismic wavelet support in curvelet domain

Most traditional ground roll separation methods utilize only the difference in geometric characteristics between the ground roll and the refl ection wave to separate them.When the geometric characteristics of data are complex,these methods often lead to damage of the reflection wave or incompletely suppress the ground roll.To solve this problem,we proposed a novel ground roll separation method via threshold filtering and constraint of seismic wavelet support in the curvelet domain;this method is called the TFWS method.First,curvelet threshold fi ltering(CTF)is performed by using the diff erence of the curvelet coeffi cient of the refl ection wave and the ground roll in the location,scale,and slope of their events to eliminate most of the ground roll.Second,the degree of the local damaged signal or the local residual noise is estimated as the local weighting coeffi cient.Under the constraints of seismic wavelet and local weighting coeffi cient,the L1 norm of the refl ection coeffi cient is minimized in the curvelet domain to recover the damaged refl ection wave and attenuate the residual noise.The local weighting coeffi cient in this paper is obtained by calculating the local correlation coeffi cient between the high-pass fi ltering result and the CFT result.We applied the TFWS method to simulate and fi eld data and compared its performance with that of frequency and wavenumber filtering and the CFT method.Results show that the TFWS method can attenuate not only linear ground roll,aliased ground roll,and nonlinear noise but also strong noise with a slope close to the refl ection events.

基于ICEEMDAN-SSA-Wavelet的声发射信号降噪研究

针对粘接件声发射(AE)信号含有噪声分量难以滤除的问题,提出一种改进ICEEMDAN的方法。该方法首先使用ICEEMDAN分解原始AE信号,并通过相关系数和能量差值的方法筛选出低频分量和高频分量;运用麻雀优化算法(SSA)优化后的改进小波阈值去噪算法对其进行去噪;最后将保留的低频分量和去噪后的高频分量重构成一个新的信号,通过实验数据对比和分析评估降噪效果。实验结果表明,相较于改进小波阈值去噪和ICEEMDAN去噪,文中提出的方法对金属与非金属粘接件AE信号的降噪效果更好,能够保护原始信号的频域信息,进而提高脱粘检测精度。

A novel non-separation opening scheme of front cover for rocket launch canister

In view of that existing opening technologies of front cover for rocket launch canister have disadvantages such as causing damage on the ground equipment,not being reused and easily broken.A novel reusable non-separation spring-driven opening scheme is proposed to achieve rapid and reliable opening of the front cover.The mathematical model of the opening process of the front cover is established by the rigid body dynamics theory.To establish a response surface model to optimize the opening scheme,three main influencing factors of the opening process are obtained through the designed experiments,including the pre-compression,the stiffness of the thrust spring,and the thrust spring force arm length.In addition,the prescribed kinematic law was taken as constraint,and the smaller thrust spring preliminary pressure and angular velocity was taken as optimization expectations.The results show that the opening scheme meets the design requirements on opening process well.It also shows that the optimized scheme can reduce the kinetic energy of the front cover,and the impact on the canister effectively,achieving a reliable and rapid opening of the front cover.

Preparation of Porous Silicone Resin Sheet with Phase Inversion in Parallel with Non Solvent Induced Phase Separation and Application to Hollow Particle Formation

We have tried to prepare the porous silicone resin sheet with the phase inversion method in parallel with the non solvent induced phase separation method. In the experiment, ethyl acetate and water were adopted as a good solvent and a poor solvent for silicone resin, respectively and ethyl alcohol as an amphiphilic solvent was used to increase the solubility of ethyl acetate in water and decrease the interfacial tension by mass transfer from water to ethyl acetate. The concentration of silicone resin in ethyl acetate and the oil soluble surfactant species were changed. Increasing the concentration of silicone resin could depress coalescence between the water droplets in the (W/O) dispersion and increase the porosity and pore number density of silicone resin sheet. Span 80 among the oil soluble surfactant species made the porosity and pore number density larger. The effect of physical proparties of liquids concerned on the porosity and pore number density was discussed on the basis of dispersing behavior of liquid droplets in the liquid-liquid dispersion. The hollow silicone resin particles could be prepared by applying the preparation method presented here.

Multi-scale separation of aeromagnetic abnormality based on dual-tree complex wavelet

Bit-field separation is an important part of gravity and magnetic data processing.In order to extract different levels of anomaly information better,this paper introduces the dual-tree complex wavelet multi-scale separation to the processing of bit-field data firstly and uses the geological model of different buried depth to ve-rify its feasibility.Finally,the dual-tree complex wavelet is applied to the aeromagnetic anomaly in Jinchuan copper nickel mining area.The results show that the method can effectively separate the anomaly information of different scales and analyze the output results with relevant geological data.

Novel wavelet-homotopy Galerkin technique for analysis of lid-driven cavity flow and heat transfer with non-uniform boundary conditions

In this paper, a brand-new wavelet-homotopy Galerkin technique is developed to solve nonlinear ordinary or partial differential equations. Before this investigation,few studies have been done for handling nonlinear problems with non-uniform boundary conditions by means of the wavelet Galerkin technique, especially in the field of fluid mechanics and heat transfer. The lid-driven cavity flow and heat transfer are illustrated as a typical example to verify the validity and correctness of this proposed technique. The cavity is subject to the upper and lower walls’ motions in the same or opposite directions.The inclined angle of the square cavity is from 0 to π/2. Four different modes including uniform, linear, exponential, and sinusoidal heating are considered on the top and bottom walls, respectively, while the left and right walls are thermally isolated and stationary.A parametric analysis of heating distribution between upper and lower walls including the amplitude ratio from 0 to 1 and the phase deviation from 0 to 2π is conducted. The governing equations are non-dimensionalized in terms of the stream function-vorticity formulation and the temperature distribution function and then solved analytically subject to various boundary conditions. Comparisons with previous publications are given,showing high efficiency and great feasibility of the proposed technique.

Nonnegative matrix factorization with Log Gabor wavelets for image representation and classification

Many problems in image representation and classification involve some form of dimensionality reduction. Nonnegative matrix factorization （NMF） is a recently proposed unsupervised procedure for learning spatially localized, partsbased subspace representation of objects. An improvement of the classical NMF by combining with Log-Gabor wavelets to enhance its part-based learning ability is presented. The new method with principal component analysis （PCA） and locally linear embedding （LIE） proposed recently in Science are compared. Finally, the new method to several real world datasets and achieve good performance in representation and classification is applied.

Separation of non-denatured proteins using semi-crosslinked polyacrylamide capillary gel electrophoresis

This work presents an approach to build a high-performance, low-viscous and replaceable separation matrix, semi-crosslinked polyacrylamide （semi-CPA） capillary gel electrophoresis. Non- denatured basic proteins, such as lysozyme, cytochrome C, ribonuclease A and trypsin were separa- ted. The impacts of monomer and cross-linker concentrations on protein separation were studied, and the ability of dynamic capillary inner wall coating was demonstrated. The UV absorption interfer- ence by semi-CPA gel matrix was successfully overcome by a partial filling technique, which results in sensitivity 20 times higher than other protein separation method. The excellent separation ability, reproducibility and dynamic coating ability made semi-CPA an ideal separation media in both capillar- y electrophoresis and microfluidic chip separation scheme.

Metal Salt and Non-Electrolyte Separation by Means of Dialysis through the Composite Membranes

To separate salts of metals and non-electrolytes, the approach of dialysis through the composite membranes (CMs) is proposed. CM is a combination of cation and anion exchange areas. In the composite membrane, cations and anions are transferred through the respective exchange areas simultaneously without violation of macroscopic electro-neutrality. This provides a better transfer of salts than conventional ion exchange membranes (IEMs). The dialysis of the ethylene glycol aqueous salt solutions through the CMs was investigated. We have shown that the transport of salts through the composite membranes is more intensive (unlike IEM providing no transfer of salts from weakly mineralized aqueous solutions due to the Donnan exclusion) and the ethylene glycol transfer is not very significant, that is the basis of effective separation. The possibility to use of composite membranes for metal salt and other electrolyte separation is discussed.

Enhanced Fourier Transform Using Wavelet Packet Decomposition

Many domains, including communication, signal processing, and image processing, use the Fourier Transform as a mathematical tool for signal analysis. Although it can analyze signals with steady and transitory properties, it has limits. The Wavelet Packet Decomposition (WPD) is a novel technique that we suggest in this study as a way to improve the Fourier Transform and get beyond these drawbacks. In this experiment, we specifically considered the utilization of Daubechies level 4 for the wavelet transformation. The choice of Daubechies level 4 was motivated by several reasons. Daubechies wavelets are known for their compact support, orthogonality, and good time-frequency localization. By choosing Daubechies level 4, we aimed to strike a balance between preserving important transient information and avoiding excessive noise or oversmoothing in the transformed signal. Then we compared the outcomes of our suggested approach to the conventional Fourier Transform using a non-stationary signal. The findings demonstrated that the suggested method offered a more accurate representation of non-stationary and transient signals in the frequency domain. Our method precisely showed a 12% reduction in MSE and a 3% rise in PSNR for the standard Fourier transform, as well as a 35% decrease in MSE and an 8% increase in PSNR for voice signals when compared to the traditional wavelet packet decomposition method.

Shannon小波混沌神经网络及其TSP(城市旅行商)问题的求解

混沌神经网络已经被证明是解决组合优化问题的有效工具.针对混沌神经网络的单调的激励函数,通过引入Shannon小波和Sigmoid函数加和组成的非单调激励函数,提出了一种新型的暂态混沌神经元模型.给出了该混沌神经元的倒分岔图和最大Lyapunov指数时间演化图,分析了其动力学特性.基于该模型,构造了一种暂态混沌神经网络.并将其应用于函数优化和组合优化问题.通过经典的10城市TSP验证了该暂态混沌神经网络的有效性.

结合血管拓扑和Non-local CNN-GCN的肺动静脉分离方法

为解决肺CT影像中动、静脉血管的自动分离、精确提取问题,提出一种结合血管拓扑与连接非局部卷积神经网络和图卷积网络(Non-local CNN-GCN)的肺动静脉自动分离方法.首先,使用尺度空间粒子采样方法将血管树表示为一组中心线体素粒子;然后,根据粒子体素及其相邻体素构造拓扑图,并从拓扑图中提取只包含动脉或静脉分支的拓扑子图;最后,基于拓扑图训练Non-local CNN-GCN分类器,并利用拓扑子图来改进分类结果.在16例CT影像数据上进行训练和测试,结果显示,本方法能有效去除动静脉黏连点,分类平均准确度达94%.

基于Wigner-Ville分布和Wavelet时间尺度的飞机非平稳抖杆背景声分析

飞机黑匣子中舱音记录器记录的舱音主要是由话语音、噪声和具有不同物理意义的背景声组成的复杂混合体,增加了舱音分析的难度,用传统的辨听和分析方法难以解决舱音特征等技术难题。以舱音记录器记录的抖杆背景声为例,针对传统Fourier变换的一次型时间/频率和短时Fourier变换的二次型能量在舱音辨听、分析的不足,将Wigner-Ville时频和Wavelet时间尺度引入到飞机舱音分析中,扩展了舱音分析方法。对比分析结果表明:平滑伪Wigner-Ville分布(SPWVD)及其时频分布的重排具有高的时频分辨率、时频凝聚性,而且没有交叉项干扰,适合分析非平稳舱音背景声;同时,离散小波变换的时间尺度分析技术确定抖杆背景声基频频率及位置,为获得非平稳舱音背景声特征提供一种值得借鉴的途径。