对协方差函数的意义和应用做了讨论。
Significance and application of covariance functions are discussed.
由协方差函数的统计特征，可给出核函数的参数估计。
According to the statistic characteristic of covariance function, parameter estimation can be given for kernel function.
最后讨论了零内尺度模型下归一化强度协方差函数随湍流强度的变化情况。
Finally, the variation of the normalized covariance function of intensity with different strengths of turbulence under zero inner model is discussed in detail.
结果表明，协方差函数可以反映出体重在连续的年龄尺度上的遗传和表型变异。
The results showed that covariance functions could reflect the changes of genetic and phenotypic variation for body weight in continuous scale of age.
讨论了不同湍流强度、波长和传播距离对强度协方差函数和归一化强度方差的影响。
The influence of turbulence strength, wavelength and propagating distance on the covariance function and normalized variance of the scattering intensity is discussed.
通过长程情况下的马尔科夫近似，得到了互相关函数，对数振幅和相位协方差函数。
According to the Markov approximation under a long haul condition, we get the inter-correlation function, log-amplitude and phase covariance function.
结果表明，窝效应只能反映部分永久环境效应，估计协方差函数时需配合全阶多项式；
The results showed that litter effect only reflected part of the permanent environmental effects and full order fit was necessary to estimate the covariance functions.
本文通过对航空重力测量数据的分析，建立起具有空间相关特性的空间协方差函数模型。
By the analysis of airborne gravity measurement data, this paper constructs a space covariance function model with spatial characteristics.
本文通过空间扰动位协方差函数特性，得出卫星重力梯度数据与引力位系数的相关协方差函数。
Least Squares Collocation is used to educe the function that can directly compute gravitational potential coefficient by satellite gravity gradiometry data.
其根据多普勒雷达径向速度资料的特点，水平方向采用非各向同性的单变量背景场误差协方差函数。
As for the characters of the Doppler Radar radial velocity data, the single variable background error covariance is used. And the anisotropic covariance function is adopted in the horizontal plane.
本文提出了一种新型的并行三维纹理综合算法，即：基于图象的自协方差函数，均值和方差的纹理综合方法。
A new 3-d texture synthetic method is proposed in this paper, i. e., the parallel 3-d texture synthetic algorithm based on autocorrelation parameters, the mean and the variance of texture images.
与传统的方法相比，无需对随机信号的协方差函数的功率谱进行分解，不受限于协方差函数的功率谱是否为有理式。
Compared with the traditional method, it does not need disassembling the power spectral densities of stochastic signal covariance and is not bound to if the power spectral densities are rational.
表明： 基于氨基酸组成和有偏自协方差函数为特征矢量的BP神经网络预测蛋白质二级结构含量的方法可有效提高预测精度。
It is shown that the BP neural network method combined with the amino-acid composition and the biased auto-covariance function features could effectively improve the prediction accuracy.
其中，对于观测向量协方差阵的谱分解估计，我们很容易得到它在一些损失下的风险函数。
Thereinto, for the spectral decomposition estimate of the covariance matrix , we can gain the risk functions under some losses.
提出利用高斯随机分布的密度函数设置稀疏阵列，在稀疏阵列得到的协方差矩阵经扩展后，增益有了明显的提高。
This paper points out that we can use the density function of the Gaussian distribution to set a thinned array, extending the covariance matrix will advance the gain clearly.
另一方面又利用矩阵理论等推导出含有二次项的非线性函数的协方差传播公式的矩阵形式。
On the other hand, it also derives the covariance propagation formula of non-linear function containing quadratic term in matrix form using the matrix theory etc.
也为采用某些数学方法(如协方差描述函数法)进一步研究火箭发射系统的统计性能作了必备的前提工作。
So the study makes an easy way for launching research of MLRS and prepares for applying some mathematics methods (such as CADET) in the statistic field of MLRS.
提出利用高斯随机分布的密度函数设置稀疏阵列，稀疏阵列得到的协方差矩阵经扩展后，增益会有明显的提高。
This paper points out that a thinned array can be set with the density function of the Gaussian distribution, after the extending of covariance matrix the gain can be increased clearly.
本文的目的在于，对于线性平稳时间序列的样本、自协方差、自相关和偏相关函数的渐近性质，给出一个比较系统的描述。
The aim of this paper is to give a systematic account of asymptotic properties of the sample autocovariance, autocorrelation and partial autocorrelation functions of linear stationary time series.
对于不同的相关函数模型对背景误差协方差的拟合也做了相关分析对比。
Different correlation functions fitting background error covariance are investigated and the results are given.
提出了一种基于信息效能函数的传感器管理最优决策模型，分析了以预测误差协方差为分配效能的分配算法的最优性。
An optimal decision model of sensor management based on information efficacy function is proposed. The optimality of the allocation algorithm based on predicting covariance is analyzed.
这种方法通过由相关函数抽样序列形成的协方差矩阵控制序列的相关性，适用于仿真具有不同概率密度函数的各种有限长相关的随机序列。
This method can control the series correlation via covariance matrix that was formed from correlation series and is suit for simulating finite correlated random series of different distributions.
结果双变量多水平模型可以估计各水平两个变量的方差协方差阵，据此可以计算出相关系数随协变量变化的函数式。
Results Multilevel models can present the variance covariance metrics of two dependent variables in every levels, and make out the functional expresses of correlation coefficient with covariates.
结果双变量多水平模型可以估计各水平两个变量的方差协方差阵，据此可以计算出相关系数随协变量变化的函数式。
Results Multilevel models can present the variance covariance metrics of two dependent variables in every levels, and make out the functional expresses of correlation coefficient with covariates.
应用推荐