一般线性模型下的线性充分性

讨论一般线性模型＜ｙ，ｘβＤ＞下的线性充分性和最小线性充分性，得到相应的刻一理。     

带状随机线性系统状态的估计问题

针对目前关于广义系统状态估计的研究现状 ,即几乎所有的讨论都集中在奇异矩阵为方阵的情况 ,讨论了广义逆矩阵和矩阵的奇异值分解问题。首次提出了带状随机系统的状态估计问题 ,利用矩阵的奇异分解理论给出了带状广义连续线性系统的奇异值标准形式。基于标准形式 ,利用广义逆矩阵的性质 ,在两种情况下将系统分解成两个子系统 ,通过估计子系统的状态 ,得到了该系统状态的最优预测和滤波递推方程。     

水准网平差的随机游动论解释

应用随机游动中质点的转移矩阵以及在各点逗留的期望等概念，来解释测量平差中的水准网平差问题，导出了一种能方便地计算任意两点间的最或是高差的公式和精度估算公式，算例表明，计算结果是正确的，计算过程很简便。     

加入 WTO对我国道路运输业的影响及应对之策

入世后，我国道路运输业打入国际市场，参与国际竞争的困难比较大。文章对入世后道路运输业面临的压力和机遇进行了正反两方面分析，提出了道路运输业应采取的应对措施。     

A super-resolution algorithm for spectral estimation and time series extrapolation

Recent developments in the signal processing field of electrical engineering have resulted in several frequency domain methods of extrapolating a time series. Insight gained in testing one such method, the Papoulis algorithm, has been used to suggest modifications which greatly improve its performance under most operating conditions where real data are concerned. The modified Papoulis method thus developed has been applied to electricity load forecasting over the short and medium term, as well as to world economic and energy data, to assess the cyclic structure present in each series about a trend.     

A unified view of statistical forecasting procedures

A large number of statistical forecasting procedures for univariate time series have been proposed in the literature. These range from simple methods, such as the exponentially weighted moving average, to more complex procedures such as Box–Jenkins ARIMA modelling and Harrison–Stevens Bayesian forecasting. This paper sets out to show the relationship between these various procedures by adopting a framework in which a time series model is viewed in terms of trend, seasonal and irregular components. The framework is then extended to cover models with explanatory variables. From the technical point of view the Kalman filter plays an important role in allowing an integrated treatment of these topics.     

Fitting of growth curves over time when the data are obtained from a single realization

This paper focuses on the general problem of forecasting the maximum value of a time series which by the nature of the data must approach an asymptotic value. Examples of such series include the growth of organisms, the concentration of a chemical reagent during a reaction occurring over time or the amount of a fossil fuel resource which has been discovered or produced as a function of time. The approach taken below differs from the usual models for this type of data in that it assumes that an unobserved time series is actually driving the process, and that the observed data series is a function of the unobserved process. In the case of fossil fuels the unobserved series might be a measure of the exploratory drilling, the number of production days in a given time period or even the amount of fiscal resources devoted to exploratory activities. A maximum likelihood method is developed for estimating the parameters of the process, especially the maximum S, and the covariance structure of the estimators is developed. The methodology is illustrated on an example of oil production. Finally, methods are developed for forecasting the data into the near future.     

面向脉冲噪声环境的多径信道鲁棒估计方法

在低压电力线通信(power line communication,PLC)中,由于脉冲噪声的干扰和阻抗不匹配引起的多径信道的影响,传统的在加性高斯白噪声环境下的信道估计算法运用到电力线通信时,往往效果很差,甚至得到的结果毫无意义。针对电力线通信中的脉冲噪声,采用了Middleton Class A脉冲噪声模型。通过基于高阶矩的方法对脉冲噪声的参数进行估计。根据估计得到的脉冲噪声参数,提出了新颖的、在脉冲噪声环境中鲁棒的信道估计算法。通过仿真,验证了提出的算法是可靠的,并且具有一定的灵活性。     

基于特征层融合和随机投影的行为识别算法

提出一种基于特征层融合和随机投影的行为识别算法;该方法提取视频序列的时空梯度特征和Gabor特征;然后进行特征层融合,得到分类能力更强的特征,有效地表征人体行为;同时,使用随机投影对融合后的特征进行降维;最后,为了解决主题模型参数估计迭代复杂的问题,将贝叶斯参数估计法应用于LDA(latent dirichlet allocation)主题模型中,对视频中的行为进行分类。在公开的KTH和Weizmann数据集上进行了实验,结果表明方法不仅比单一局部时空特征描述符识别性能好,而且在相同实验设置下,也优于其他基本分类器。