Wavelet denoising and feature extraction of seismic signal for footstep detection

Seismic sensors are widely used to detect moving target in ground sensor networks. Footstep detection is very important for security surveillance and other applications. Because of non-stationary characteristic of seismic signal and complex environment conditions, footstep detection is a very challenging problem. A novel wavelet denoising method based on singular value decomposition is used to solve these problems. The signal-to-noise ratio (SNR) of raw footstep signal is greatly improved using this strategy. The feature extraction method is also discussed after denosing procedure. Comparing, with kurtosis statistic feature, the wavelet energy feature is more promising for seismic footstep detection, especially in a long distance surveillance.

基于MATLAB小波分析的软地基加固探地雷达检测处理方法

从小波分解出发，利用多尺度分解对控地雷达波进行分析，并基于MATLAB的小波分析对深圳市公安局某工地强夯土石墩地基的检测结果进行处理，提出了具体的处理方法。

小波分析和人工神经网络在生态学研究中的应用

墨西哥帽子小波和Morlet小波在生态格局分析中的应用 本研究采用Monte Carlo方法，探讨了对小波分析的格局进行统计显著性检验的普遍方法。为了更好利用小波分析和了解两个常用小波--墨西哥帽子小波和Morlet小波在格局分析中的优缺点，用生态学研究中常见的4个模拟格局和东灵山辽东栎林的样带数据对这两个常用小波的特性进行了分析和比较。研究结果表明：墨西哥帽子小波能较好地分析样带中的斑块和间隙以及它们的位置信息，Morlet小波能较好地分析样带中尺度及其位置信息。不同的小波通常在尺度分析和斑块和间隙分析之间有平衡，所以最好的方法是结合两种小波的优点。小波分析在处理生态数据时，受所使用小波本身特性的制约。用墨西哥帽子小波进行格局分析时，小波能量谱的等值图上不同格局所对应的峰有可能相互重叠，当所分析的多尺度格局的规模差别不大时，所分析格局规模所对应的峰可能相互融合。这些小波能量谱的等值图上不同格局所对应峰的重叠和融合可能会导致格局分析量图上多个峰的相互融合和屏蔽。所以在使用小波分析做格局研究时，也应尽可能地结合小波能量谱和量图上的信息，以得到较全面和精确的格局分析信息。本研究的结论能为小波分析的应用提供指导。 应用二维小波分析对暖温带阔叶林辽东栎更新格局的研究 本研究介绍了一种二维网格空间数据分析方法一二维小波分析。该方法不仅能分析格局的等级结构，而且也能得到所分析结构的位置信息。小波系数等值图上不同格局规模的斑块和间隙可直接和不同尺度上的生态过程和生境条件相联系。小波方差从二维小波分析导出，小波方差可将四维的小波系数降至二维的小波方差函数，并量化所分析格局规模对整个格局的贡献。本研究用三个模拟格局分析了二维小波的特性及二维墨西哥帽子小波和Halo小波的特性。因为自身的特性，Halo小波比墨西哥帽子小波能提供更高的分辨率。本研究也将二维小波分析应用于暖温带阔叶林的辽东栎更新格局研究中，分析的结果表明：辽东栎的更新发生在辽东栎成树斑块和林窗斑块重叠区域。 用交互验证和独立验证来测试神人工经网络模拟水稻分檗动态的泛化能力 人工神经网络不是基于对所模拟过程的理解，而是依赖于对所分析数据的内部结构。所以人工神经网络通常被认为是经验模型而不能外推，而且在训练数据和验证数据的范围之外肯定不能精确地预测所模拟的过程。本研究通过对交互验证和独立验证神经网络模型性能的比较，测试了神经网络模型在预测水稻分檗动态方面的泛化能力。同时，也对几种提高神经网络泛化能力的技术进行了比较。研究结果表明：在训练数据的变量范围内，神经网络在预测水稻分檗动态方面具有泛化能力。较少的训练数据样本导致了对训练数据过度吻合的和不具泛化能力的神经网络。要能使神经网络在预测水稻分檗动态方面具有泛化能力，训练数据的样本量至少9倍于神经网络的权值数目。当神经网络有多个输入变量且训练数据不足以保证神经网络的泛化能力时，建议在训练之前，采用主成分分析、对应分析及类似技术压缩输入变量的个数。在压缩输入变量的个数之后，如果训练数据的样本量还不足以保证神经网络的泛化能力，就应采用提高神经网络泛化能力的技木，如：jittering和强制训练停止等，特别是神经网络与机理模型的复合模型。因为神经网络的泛化能力问题具有普遍性，所以我们的研究结论不只是适用于水稻分檗动态的预测，也适用于其它的农业和生态神经网络模型。

延河流域径流和输沙周期变化特征的小波分析

用延河流域甘谷驿站1953—2000年实测月径流深和输沙模数数据和1957—2000年月均面降雨数据,采用小波多尺度方法,分析延河流域主要水文要素的周期及其变化特征。结果表明:延河流域降雨、径流和输沙存在着显著的周期,且其大小非常接近,为3.0、6.5、13.0和23.0 a。降雨、径流和输沙主周期分别是3.0、23.0和23.0 a。延河流域20世纪70年代后降雨量年际分配趋于均匀。3 a对应的小尺度上,70—80年代初3个水文序列对应曲线出现紊乱现象,与该时期大规模水利水土保持措施的修建有关。80年代后,由于水利水土保持措施减水减沙效益的削弱,3条曲线重新趋于一致。13 a对应的中尺度和23 a对应的大尺度上,均出现径流和输沙曲线不同步,或者滞后于降雨曲线现象。降水是径流和输沙周期性变化的主要外动力因子,人类活动导致的流域下垫面变化则是不可忽视的另一重要原因。

延安市近50年来降水特征及趋势变化的小波分析研究

利用Morlet小波对延安1952~2003年52年间的降水量进行了多时间尺度分析,结果表明,延安降水在3~6 a和25~32 a的时间尺度上变化较强,存在近22 a和3 a的降水周期,其中22 a的周期变化最为强烈。根据降水周期,延安年降水在2003年后的5~8 a内相对偏多,2007~2009年左右达到最大值,然后降水量开始减少,从2013年左右开始进入一个约10 a的少雨阶段。基于1952~1990年降水量资料,建立小波变换的最近邻抽样回归模型,用1990~2003年13 a的年降水量变化作为校验,13年中有11年的预测结果相对误差小于30%,说明小波分析和最近邻抽样回归模型结合可以较好地预测降水量的年际变化。

Stable isotopic compositions in Australian precipitation

Stable deuterium (delta D) and oxygen-18 (delta O-18) isotopes in 1962 to 2002 precipitation from the seven Australian stations of the Global Network of Isotopes in Precipitation (GNIP) were used to investigate isotope characteristics including temporal and spatial distributions across different regions of Australia. On the basis of 1534 samples, the local meteoric water line (LMWL) was established as delta D = 7.10 delta O-18 + 8.21. delta O-18 showed a depletion trend from north and south to central Australia (a continental effect) and from west to east. Precipitation amount effects were generally greater than temperature effects, with quadratic or logarithmic correlations describing delta/T and delta/P better than linear relationships. Nonlinear stepwise regression was used to determine the significant meteorological control factors for each station, explaining about 50% or more of the delta O-18 variations. Geographical control factors for delta O-18 were given by the relationship delta O-18 (parts per thousand) = -0.005 longitude (degrees) - 0.034 latitude (degrees)-0.003 altitude (m) - 4.753. Four different types of d-excess patterns demonstrated particular precipitation formation conditions for four major seasonal rainfall zones. Finally, wavelet coherence (WTC) between delta O-18 and SOI confirmed that the influence of ENSO decreased from east and north to west Australia.

小波消噪在振动信号处理中的应用

信噪分离是小波分析中的一个有效的信号检测方法。本文在阐述了小波分析及消噪的小波理论的基础上 ,给出了从噪声污染信号中恢复原信号的实例 ,并与傅立叶分析消噪进行了比较 ,结果表明对于非平稳振动信号小波消噪的效果明显优于傅立叶变换。

残留盆地分布综合地球物理研究——以大港地区为例

Macro-distribution of residual basins is a basic question in residual basin research，the main object of macro-distribution study is to build strata framework, compute thickness of residual strata and analyze characteristics of residual basins. With the guidance of the theory of integrated geology and geophysical research, the paper assembled series of methods and established the technical chart based on gravity and magnetic data, with restriction of geology, seismic and drilling data. Based on potential field data processing and analysis, forward and inverse computation, region potential field analysis and potential field separation, etc. it computed depth of gravity/magnetic basement and got strata framework. It had got effective results in the research of macro-distribution of residual basin research in the Dagang area. It did the wavelet transform of gravity/magnetic data with multi-kind of wavelet basis using a trou algorithm. From comparison of processing result and their spectral of wavelet analysis, up continuation and filter method, the wavelet approximation is better to fit the regional potential field, and it is an effective method to separate gravity/magnetic effect caused by deep geology bodies. The experiment of matching pursuit shows that te transform domain methods have great advantage in potential data analysis. From the integrated geophysical study of rock property study, gravity/magnetic basement inversion and fault system analysis of the Dagang area, it gets the strata framework and the thickness of pre-Cenozoic residual strata. Comprehensive study with gravity and magnetotelluric profile inversion and interpretation, three prospect plays of macro-distribution of residual basins are fingered out. It has great residual strata thickness in the northern part of Chengning Uplift and there is thrust fault in the deep zone and good up-Paleozoic hydrocarbon source rocks in this area. With integrated analysis, this area will be the most prospective hydrocarbon location of pre-Cenozoic residual basins.

砂泥岩薄互层储层预测方法与应用研究

With the deeply development of exploration and development in petroleum in China, new increasing reserves are found in old oil fields and the verge of the old ones through re-study of geological property. It is more and more important to discovery and develop thin layer or thin inter-bedded layers reservoirs. All of the targets are thin sand-shale inter-bedded reservoirs and the core technology is reservoir predictions between wells in thin sand-shale inter-bedded layers. The continuity of the thin sand-shale inter-bedded layers in space or separating and heterogeneity is the key of reservoir geology research. The seismic reflection, high resolution analysis method and inversion method to thin sand-shale inter-bedded layers are thorough discussed and deeply studied in this paper to try to find the methods and resolutions of reservoir geology research. The below is followed. 1. Based on the pre-research of other people, five models are created: the sand sphenoid body, interlay sandstone and interlay shale of the equal thickness, interlay sandstone of the equal thickness and interlay shale of the unequal thickness, interlay sandstone of the unequal thickness and interlay shale of the unequal thickness, interlay sandstone of the changing thickness in sequence and interlay shale of the changing thickness in sequence. Then the study of the forward modeling are conducted on the thin layer and thin inter-bedded layers geological characters and seismic reflections including amplitude, frequency, phase, wave shape and time-frequency responding in the domains of time and frequency. The affect of petro-physics difference of layers, single thin layer thickness, thickness of inter-bedded, layer number of inter-bedded, incident wavelet domain frequency and types, sample interval to seismic reflection characters, frequency spectrum and time-frequency respond of reflectivity is theoretically discussed. 2. Qualitatively analyzing the sedimentary rhythm of the thin inter-bedded layers in vertical orientation and computing the single layer thickness or the average thickness with the method of generalized S transform. Identifying the reflecting interface or lithology interface using the amplitude value of amplitude spectrum domain frequency. 3. Based on the seismic respond of thin sand-shale inter-bedded layers, bring out the high resolution analysis method of seismic data in thin sand-shale inter-bedded layers using wavelet analysis and the idea of affecting low and high frequency with middle frequency. Then analyzing the effect to the method and testing some wavelets in the method. This method is applied to the theoretical models and the field data. 4. Bring forward one improved very fast simulated annealing method (IVFSA) to resolve the problem nonlinearity and multi-parameters of the inversion in thin inter-bedded layers. And IVFSA is more productive and higher precision than general ways. 5. New target constrained function is used in the inversion based on the property of the inversion in thin inter-bedded layers. 6. Making the full use of geological and logging information, IVFSA and the new function are applied in the non-linear inversion to improve reservoir prediction and evaluation in thin inter-bedded formations combined with the idea of logging and seismic inversion. This method was applied to the field data and got good results.

各向异性与非线性介质中弹性波场传播特征研究

The topic of this study is about the propagation features of elastic waves in the anisotropic and nonlinear media by numerical methods with high accuracy and stability. The main achievements of this paper are as followings: Firstly, basing on the third order elastic energy formula, principle of energy conservation and circumvolved matrix method, we firstly reported the equations of non-linear elastic waves with two dimensions and three components in VTI media. Secondly, several conclusions about some numerical methods have been obtained in this paper. Namely, the minimum suitable sample stepth in space is about 1/8-1/12 of the main wavelength in order to distinctly reduce the numerical dispersion resulted from the numerical mehtod, at the same time, the higher order conventional finite difference (CFD) schemes will give little contribution to avoid the numerical solutions error accumulating with time. To get the similar accuracy with the fourth order center finite difference method, the half truncation length of SFFT should be no less than 7. The FDFCT method can present with the numerical solutions without obvious dispersion when the paprameters of FCT is suitable (we think they should be in the scope from 0.0001 to 0.07). Fortunately, the NADM method not only can reported us with the higher order accuracy solutions (higher than that of the fourth order finite difference method and lower than that of the sixth order finite difference method), but also can distinctly reduce the numerical dispersion. Thirdly, basing on the numerial and theoretical analysis, we reported such nonlinear response accumulating with time as waveform aberration, harmonic generation and resonant peak shift shown by the propagation of one- and two-dimensional non-linear elasticwaves in this paper. And then, we drew the conclusion that these nonlinear responses are controlled by the product between nonlinear strength (SN) and the amplitude of the source. At last, the modified FDFCT numerical method presented by this paper is used to model the two-dimensional non-linear elastic waves propagating in VTI media. Subsequently, the wavelet analysis and polarization are adopted to investigate and understand the numerical results. And then, we found the following principles (attention: the nonlinear strength presented by this paper is weak, the thickness of the -nonlinear media is thin (200m), the initial energy of the source is weak and the anisotropy of the media is weak too): The non-linear response shown by the elastic waves in VTI media is anisotropic too; The instantaneous main frequency sections of seismic records resulted from the media with a non-linear layer have about 1/4 to 1/2 changes of the initial main frequency of source with that resulted from the media without non-linear layer; The responses shown by the elasic waves about the anisotropy and nonlinearity have obvious mutual reformation, namely, the non-linear response will be stronger in some directions because of the anisotropy and the anisotropic strength shown by the elastic waves will be stronger when the media is nonlinear.

Evaluation of a combined wavelet and a combined principal component analysis classification system for BCG diagnostic problem

Heart disease is one of the main factor causing death in the developed countries. Over several decades, variety of electronic and computer technology have been developed to assist clinical practices for cardiac performance monitoring and heart disease diagnosis. Among these methods, Ballistocardiography (BCG) has an interesting feature that no electrodes are needed to be attached to the body during the measurement. Thus, it is provides a potential application to asses the patients heart condition in the home. In this paper, a comparison is made for two neural networks based BCG signal classification models. One system uses a principal component analysis (PCA) method, and the other a discrete wavelet transform, to reduce the input dimensionality. It is indicated that the combined wavelet transform and neural network has a more reliable performance than the combined PCA and neural network system. Moreover, the wavelet transform requires no prior knowledge of the statistical distribution of data samples and the computation complexity and training time are reduced.

Time-Frequency Analysis of Nonlinear Systems: The Skeleton Linear Model and the Skeleton Curves

The joint time-frequency analysis method is adopted to study the nonlinear behavior varying with the instantaneous response for a class of S.D.O.F nonlinear system. A time-frequency masking operator, together with the conception of effective time-frequency region of the asymptotic signal are defined here. Based on these mathematical foundations, a so-called skeleton linear model (SLM) is constructed which has similar nonlinear characteristics with the nonlinear system. Two skeleton curves are deduced which can indicate the stiffness and damping in the nonlinear system. The relationship between the SLM and the nonlinear system, both parameters and solutions, is clarified. Based on this work a new identification technique of nonlinear systems using the nonstationary vibration data will be proposed through time-frequency filtering technique and wavelet transform in the following paper.

Comparison of wavelet transform methods - The compression of IR spectral data

In this paper, an introduction of wavelet transform and multi-resolution analysis is presented. We describe three data compression methods based on wavelet transform for spectral information,and by using the multi-resolution analysis, we compressed spectral data by Daubechies's compactly supported orthogonal wavelet and orthogonal cubic B-splines wavelet, Using orthogonal cubic B-splines wavelet and coefficients of sharpening signal are set to zero, only very few large coefficients are stored, and a favourable data compression can be achieved.