欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
随机信号协方差函数的拟合和确定是拟合推估的关键。在常规协方差函数拟合时,通常假定随机信号为具有各向同性的随机过程,而事实上各向异性更具有普遍性。结合方差在不同方向的误差分量表达式,给出了各向异性协方差函数的拟合方法,利用由方差分量估计构建的自适应因子调节观测随机误差与信号对模型参数估计的贡献,以减弱观测误差和随机信号先验模型不确定而带来的影响,并将其应用于InSAR监测缺失数据填补中。计算结果表明,拟合推估具有较好的缺失数据填补能力,应用基于各向异性协方差函数的自适应拟合推估,其填补精度得到进一步的改善。
英文摘要：
Covariance function fitting for stochastic signal is a key problem for collocation. In the process of covariance function estimation, we often assume stochastic signals with the characteristic of isotropic, but anisotropic is more generally. Combined the expressions of error component in different directions, a method of covariance function fitting based on anisotropic is given. Further, adaptive collocation, which constructed by variance component estimations can adjust the contribution to model parameters by observation errors and stochastic signals and weaken the affect leading by their uncer- tainty, is proposed and applied in missing InSAR data fitting. A practical example shows that collocation can well supply the missing data, and adaptive collocation based on anisotropic covariance function can give higher accuracy in missing data fitting.
同期刊论文项目
同项目期刊论文
期刊信息
位置：