自回归过程的小波神经网络逼近算法

将小波变换原理与非线性时间序列逼近问题相结合,利用模型分析方法,对一阶非线性自回归过程（ＮＬＡＲ（１））的小波神经网络逼近问题进行了研究,构造出了逼近函数的具体表达式,并给出ＮＬＡＲ（１）过程所逼近的误差限。     

一种估计自回归模型参数的新方法

本文应用最小二乘法给出了一种无需事先计算自相关函数而直接估计自回归模型参数的迭代方法。     

基于小波方法的空间风速场模拟

为克服随时间变化模型基于适应过滤器和窗口估计方法中窗口长度不宜过大,产生的短时时间序列模拟准确性不够高的缺点,应用小波分解方法,结合线性滤波器法的向量过程自回归(AR)模型,给出了模拟空间脉动风场的一种新方法.该方法对于AR模型自回归系数在空间上进行小波扩展,采用最小二乘法来估计AR模型的自回归系数,并给出了该方法模拟空间风速场的实现步骤.将该方法应用于一空间结构的风速场模拟,并给出了模拟结果与目标值的对比,以及与向量过程AR模型模拟结果的对比,结果证实该方法可以减少风速时程分析在频域上的信息损失,对短时时间序列模拟具有较高的准确性,并具有较高的计算效率.     

一类自相似Volterra Gauss过程的遍历变换

提出了一类X的在坐标空间上的保测变换,推广了这一结果,证明了这些保测变换是遍历的,得出了一类Volterra型的自相似Gauss过程的遍历变换.     

自回归模型参数的普通最小二乘估计

将线性回归模型参数的普通最小二乘估计推广到12种自回归模型的参数估计。     

自回归模型参数的加权最小二乘估计

将线性回归模型参数的加权最小二乘估计推广到12种自回归模型的参数估计。     

自相似网络流量Hurst指数估计算法

Hurst参数是表征网络业务自相似性的重要参数,在一定的观察时间内对突发业务的Hurst参数进行快速、准确地估计是高速宽带网络(如ATM)实施流量控制和缓冲资源分配的前提。提出一种基于DFGN模型和Haar小波的Hurst参数估计方法。仿真生成的DFBM和真实自相似网络业务数据的计算结果均表明,所述方法提高了Hurst参数估计的效率和准确性,比传统方法具有更好的性能。     

基于小波变换的分形随机信号希尔伯特变换的波形估计

非平稳1/f类分形随机过程是一类重要的弱自相似随机过程,其典型例子是分形布朗运动,作者考察了弱自相似过程的希尔伯特变换,证明了希尔伯特变换具有弱自相似不变性,进而利用分形布朗运动的希尔伯特变换BH（t)小波系数能量向低频集中的特性,采用最优门限方法实现BH（t)的波形估计。     

自回归过程逆自相关函数的估计

这篇文章使用Kanto公式[8]给出了自回归过程逆自相关函数的一种直接估计方法。并且证明了这种估计的强相容性和渐近正态性     

网络流量的自相似特性

利用小波技术对网络流量进行特性分析,将网络流量过程分解成不同尺度下的小波系数(细节)和尺度系数(背景),从而刻画出网络流量的自相似特性.     

Some Results about the Sample Path Properties of Markov Processes with Independent Self-Similar Components

This paper considers a special class of operator self-similar processes Markov processes {X（t）, t≥0} with independent self-similar components, that is, X （ t ） =（X^1（t）,…,X^d（t））, where {X^i（t）,t≥0}, i=1,2,…,d are d independent real valued self-similar Markov processes. By means of Brel-Cantelli lemma, we give two results about asymptotic property as t→∞ of sample paths for two special classes of Markov processes with independent self-similar components.     

The exact measure functions of the images for a class of self-similar processes

Let {ie1-1} be a self-similar Markov process on (0,∞) with non-decreasing path. The exact Hausdorff and Packing measure functions of the imageX([0,t]) are obtained. Foundation item: Supported by the National Natural Science Foundation of China Biography: HUANG Li-hu(1972-), male, Ph. D graduate student, research interest is in stochastic processes.     

Some fractal properties of a class of self-similar Markov processes

LetX=(Ω,F,F _t ,X _t , θ _t ,P ^x be a self-similar Markov process on (0, ∞). The exact Hausdorff measure function of the level sets are obtained. An appropriate condition is given under which the self-similar Markov process corresponds to a stable process, and some more fractal properties of the sample path ofX are obtained in this case. Supported by the National Natural Science Foundation of China Huang Lihu: born in March 1972, Ph. D graduate student     

小波变换及其数字图像处理的应用

简要介绍了小波变换方法 ,对小波分析在数字图像处理占的一些应用进行了简要讨论 ,并对图像的恢复和增强、图像分割等应用进行了一些有意义的尝试     

离散子波变换计算子波级数变换的预滤波器结构

基于子波空间采样定理，提出了两种离散子波变换计算子波级数变换的预滤波器结构，消除了Ｓｈｅｎｓａ算法中形成预滤波器的积分运算，并分析了算法结构的准确性。数值计算实例验证了结构的有效性，最后讨论了用离散子波变换计算了子波级数变换这一公开问题。     

离散点云数据的小波变换处理算法

将离散点云数据表示成适合用作小波变换的形式,提出了一种基于尺度的离散点云数据的特征识别算法,在此基础上给出了具体的基于尺度的二维和三维离散点云的小波分解算法,最后引入实例对二维离散点云的小波分解算法进行分析,实验结果表明算法达到了对点云数据的按尺度特征分解的目的.通过提出的算法,将离散点云数据按照尺度进行分解并提取出不同的特征成分,这样可以根据后期可视化显示的不同要求,将小波变换分解后的数据进行进一步的处理.     

基于SV模型的VaR Monte Carlo模拟

文章以上证综合指数为研究对象讨论了Monte Carlo模拟法计算VaR,分别利用SV-VaR模型和Monte Carlo-SV-VaR模型计算VaR值,并比较了这2个模型的精确度,从而为上证综合指数的风险度量提供一个参照。     

Selecivity estimation using compressed spatial information

Spatial selectivity estimation is one of the essential studies to get query responses rapidly and accurately with the limitation of memory space. Currently, there exist several spatial selectivity estimation techniques such as random sampling, histogram, and parametric. Especially，Cumulative Density Histogram guarantees accurate estimation for rectangle object which has multiple-count problem. However，it requires large memory space because of retaining four sub-histograms for spatial data. Therefore in this paper， we propose a new technique Cumulative Density Wavelet Histogram，called CDWH，which is the combination of Cumulative Density Histogram and Haar Wavelet Transform，a compressed technique. The proposed method simultaneously takes full advantage of their strong points，high accuracy provided by the former and economization of memory space supported by the latter. Consequently，our technique is able to support estimates with relatively low error and retain similar estimates even if memory space is small.     

Selectivity estimation using compressed spatial information

Spatial selectivity estimation is one of the essential studies to get query responses rapidly and accurately with the limitation of memory space. Currently, there exist several spatial selectivity estimation techniques such as random sampling, histogram, and parametric. Especially, Cumulative Density Histogram guarantees accurate estimation for rectangle object which has multiple-count problem. However, it requires large memory space because of retaining four sub-histograms for spatial data. Therefore in this paper,we propose a new technique Cumulative Density Wavelet Histogram, called CDWH, which is the combination of Cumulative Density Histogram and Haar Wavelet Transform, a compressed technique. The proposed method simultaneously takes full advantage of their strong points, high accuracy provided by the former and economization of memory space supported by the latter. Consequently, our technique is able to support estimates with relatively low error and retain similar estimates even if memory space is small.     

Selectivity estimation using compressed spatial information

Spatial selectivity estimation is one of the essential studies to get query responses rapidly and accurately with the limitation of memory space. Currently, there exist several spatial selectivity estimation techniques such as random sampling, histogram, and parametric. Especially, Cumulative Density Histogram guarantees accurate estimation for rectangle object which has multiple-count problem. However, it requires large memory space because of retaining four sub-histograms for spatial data. Therefore in this paper, we propose a new technique Cumulative Density Wavelet Histogram, called CDWH, which is the combination of Cumulative Density Histogram and Haar Wavelet Transform, a compressed technique. The proposed method simultaneously takes full advantage of their strong points, high accuracy provided by the former and economization of memory space supported by the latter. Consequently, our technique is able to support estimates with relatively low error and retain similar estimates even if memory space is small.