一种高精度的非平稳随机信号实时建模算法研究

针对传统的非平稳随机信号时变参数筹分模型算法的不足,论文把非平稳随机信号时变参数差分模型与一种带自适应遗忘因子的RLS算法结合起来,形成一种有效的非平稳随机信号实时建模算法,仿真分析表明该算法不仅能获得快的收敛速度而且能获得高的建模精度。     

基于聚类算法实现信号盲分类

本文解决了信号处理、工业控制等领域存在的非平稳信号盲分类问题。在聚类中广泛应用的K-Means算法及其它基于中心的聚类算法有两个共同的缺陷-需要预先确定类数目且随机初始化中心引起性能不稳定。本文提出的算法较好地解决了这两个问题，提高了算法稳定性，实现了非平稳信号盲分类。提取非平稳信号的小波系数作为聚类的样本空间，分析聚类结果的统计偏差以估计类的数目，采用调和均值准则进行分类。最后给出的仿真结果表明本文提出的方法较传统的K-Means算法明显降低分类错误率。     

金融时间序列分析与建模

介绍了金融时间序列分析与建模的主要理论、方法,着重论述了近20多年发展起来的非平稳时间序列的建模方法及工具,指出进一步的研究方向.最后,对上证综合指数这一金融时间序列进行了实证分析.     

探地雷达回波信号的特征提取和分类

探地雷达的回波信号具有非平稳特征.为实现非平稳信号的检测和分类,采用了二进小波变换、主分量分析与Fisher线性判别分析和前馈多层感知器分类器分别对探地雷达回波信号进行小波分解、去相关与特征选择和分类.实验结果表明,该算法具有很好的分类效果.     

An inverse problem for Maxwell's equations in isotropic and non-stationary media

In this article, we consider Maxwell's equations in an isotropic, inhomogeneous and non-stationary medium. We discuss an inverse problem of determining the t-independent components of the coefficients ?, μ in the constitutive relations from a finite number of interior measurements. We prove a Lipschitz stability estimate for the inverse problem by applying the argument on the basis of Carleman estimate.     

Tunnelling effect of the non-stationary Kerr black hole

Extending Parikh and Wilczek＇s work to the non-stationary black hole, we study the Hawking radiation of the non-stationary Kerr black hole by the Hamilto-Jacobi method. The result shows that the radiation spectrum is not purely thermal and the tunnelling probability is related to the change of Bekenstein Hawking entropy, which gives a correction to the Hawking thermal radiation of the black hole.     

基于支持向量机方法对非平稳时间序列的预测

目前有关非平稳复杂系统及其在预测中的应用研究是一个较少被人理解并有重大科学意义的前瞻性研究课题.在大气运动中,气候正是一个典型的非平稳系统,但是现有的气候预测理论,包括统计预测理论和非线性预测理论,几乎都无一例外地建立在平稳性假定的基础之上,这有悖于气候过程的基本性质,它有可能是导致气候预测水平低下的重要的理论原因.因此以分析如何降低时间序列非平稳程度作为切入点来研究短期气候预测问题有着重要的理论意义.利用基于“升维”思想的支持向量机方法对时变控制参数条件下Lorenz系统产生的非平稳时间序列以及来自实际     

Testing for Unit Roots in a Nearly Nonstationary Spatial Autoregressive Process

The limiting distribution of the normalized periodogram ordinate is used to test for unit roots in the first-order autoregressive model _st= _s-1,t+_s,t-1- _s-1,t-1+_st. Moreover, for the sequence _n = e ^c/n , _n = e ^d/n of local Pitman-type alternatives, the limiting distribution of the normalized periodogram ordinate is shown to be a linear combination of two independent chi-square random variables whose coefficients depend on c and d. This result is used to tabulate the asymptotic power of a test for various values of c and d. A comparison is made between the periodogram test and a spatial domain test.     

Optimal Control of Non-stationary Differential Linear Repetitive Processes

Differential repetitive processes are a distinct class of continuous-discrete 2D linear systems of both systems theoretic and applications interest. The feature which makes them distinct from other classes of such systems is the fact that information propagation in one of the two independent directions only occurs over a finite interval. Applications areas include iterative learning control and iterative solution algorithms for classes of dynamic nonlinear optimal control problems based on the maximum principle, and the modelling of numerous industrial processes such as metal rolling, and long-wall cutting etc. The new results in is paper solve a general optimal problem in the presence of non-stationary dynamics.