连续观测系统的平差模型与有色噪声补偿

现代测量技术多依赖于各种连续对地观测系统。有色噪声在影响参数估计的同时,还蕴含了地球科学研究所需的大量有用信息。本文讨论了连续平差模型及最小二乘解,将有限维空间中的测量平差问题推广至无穷维空间,从信号分析角度讨论了观测有色噪声的消除方法。研究表明,连续观测方程的最小二乘解可由经典最小二乘解的极限导出;有色噪声对解的影响是系统性的,然而通过合理的观测方案设计可以实现参数的无偏估计     

基于随机模型改正的有色状态噪声处理新方法

控制有色状态噪声影响除了通过状态向量扩展的方法外,也可以通过修正随机模型.提出采用多项式长除法将有色状态噪声模型展开成无穷级数,截断取其有限项,进而求取有色状态噪声方差.利用该方差对随机模型进行修正,再结合现代时间序列分析方法,构造出新的ARMA新息模型,并根据该新息模型设计状态最优滤波器.为了说明该方法的正确性和合理性,将它与标准的Kalman滤波和状态向量扩展法进行分析和比较.结果证明利用该方法能有效地控制有色状态噪声的影响.     

有色噪声作用下的抗差卡尔曼滤波

根据用GPS载波相位三差观测量进行动态定位或精密导航的需求,推导了动态噪声、观测噪声为有色噪声的抗差卡尔曼滤波公式。白噪声的抗差卡尔曼滤波是有色噪声的抗差卡尔曼滤波的特例,有色噪声的抗差卡尔曼滤波为白噪声的抗差卡尔曼滤波的推广。     

基于非线性高斯-赫尔默特模型的混合整体最小二乘估计

针对EIV模型的系数矩阵同时包含固定量和随机量的情况,通过将系数矩阵中的随机量提取出来纳入平差的随机模型,从而将EIV模型表示为非线性高斯-赫尔默特(Gauss-Herlmert,GH)模型形式,推导了混合LS-TLS(least squares-total least squares,LS-TLS)算法及其精度估计公式。算法适用于系数矩阵包含固定列、固定元素和随机元素的一般情况。模拟实例结果表明,混合LS-TLS算法与已有能够解决系数矩阵同时含固定量和随机量的结构性或加权TLS算法的估计结果一致;混合LS-TLS的估计结果统计上要优于LS或TLS估计结果。     

基于滑动L2优化准则的GPS坐标时间序列重构

GPS坐标时间序列中不仅包含白噪声,还包含闪烁噪声、随机漫步噪声等有色噪声,这些噪声将影响GPS应用的可靠性,甚至可能对一些地球物理现象做出错误的解释,因此降低GPS坐标时间序列中有色噪声的影响、提高GPS精度是一个重要和基本的问题。提出了一种滑动L₂优化估计方法(ML₂),通过选取合适的窗口建立L₂优化模型,再利用交替迭代乘子法求解每段时间序列的优化问题,并逐年滑动得到整段GPS坐标时间序列的估计。实验结果表明,ML₂方法与奇异谱分析、小波分解、滑动普通最小二乘法相比具有更好的重构效果。     

顾及系数矩阵结构性的加权总体最小二乘解算

针对加权总体最小二乘平差模型中系数矩阵具有结构性的问题,该文设计了一种顾及系数矩阵结构性的加权总体最小二乘迭代解法:首先,利用非线性最小二乘平差方法将总体最小二乘模型线性化;然后,采用结构矩阵的方法顾及系数矩阵的重复元素和常数项,通过间接平差的原理推导了顾及系数矩阵结构性的加权总体最小二乘迭代公式,可适用于加权总体最小二乘的参数估计;最后,通过算例分析并与其他算法进行比较,验证了该算法的有效性和可行性。     

广义平差的概括模型

提出了广义平差的概括模型———附有条件的最小二乘配置模型。该概括模型不仅包括滤波和推估模型 ,扩充了原有的最小二乘配置模型 ,而且经典平差模型都是它的特例。     

火山Mogi模型反演的总体最小二乘联合平差方法

针对垂直位移与水平位移的Mogi模型,提出采用总体最小二乘联合（total least squares joint,TLS-J）平差方法进行求解。该方法可同时顾及联合平差函数模型中观测向量与系数矩阵的误差项,且采用3种判别函数最小化法确定相对权比,用以权衡垂直位移与水平位移观测数据在联合求解过程中所占的比重。针对平差过程中出现的病态问题,结合L曲线法确定岭参数。通过实际算例,系统研究了总体最小二乘联合平差方法在长白山天池火山Mogi模型反演中的应用。研究结果表明,以判别函数为$\sum\limits_{i=1}^{n1}{\left| {{{\hat{\bar{e}}}}_{1i}} \right|}+\sum\limits_{j=1}^{n2}{\left| {{{\hat{\bar{e}}}}_{2j}} \right|}$的函数最小化能获得合理的压力源参数估值结果和相对权比大小,具有一定的实际参考价值。     

附不等式约束平差模型的一种快速搜索算法

大地测量中常存在一些先验不等式约束信息,充分利用它们可以保证参数解的唯一性和稳定性。然而,现有的不等式约束平差算法主要是基于优化理论,算法通常比较复杂,需要选取有效约束或建立罚函数。在最小二乘平差准则基础上,把不等式约束看成是一个可行域,借助Fisher函数在可行域中快速搜索使误差平方和达到最小的最优解,推导出了可行解为最优解的充分必要条件。建立了基于Wolfe-Powell算法的非精确快速搜索算法,从而减小了搜索算法的计算量,得到了一种新的不等式约束平差计算方法。该算法的平差准则与最小二乘平差准则一致,不需要矩阵求逆运算,可适用于维数较大的平差问题解算。     

附加傅里叶补偿项的卫星遥感影像RFM平差方法

卫星光学遥感影像的几何畸变是制约其定位精度的重要原因。采用一般系统误差补偿模型难以从根本上消除影像复杂畸变。本文在有理函数模型RFM平差方案基础上,根据傅里叶级数的逼近特性,提出用二元傅里叶多项式代替一般多项式作为系统误差补偿项,以适用符合连续条件的任意形式畸变。仿真和实际数据平差试验结果表明,本文方法能够有效补偿由于影像内外方位元素误差造成的像方定位系统误差及不同大小的畸变。在控制点充足的条件下,附加3阶傅里叶补偿项的RFM平差定位精度显著优于附加一般多项式补偿项的常规方法,其中SPOT-5异轨立体像对平差后平面和高程定位精度可分别达到3.34 m和2.48 m,QuickBird同轨立体像对平差后平面和高程定位精度分别达到0.77 m和0.54 m,均达到了子像素精度水平。二元傅里叶多项式可作为一种通用的影像系统误差补偿模型,拓展应用于航空和近景影像的畸变校正。     

复数域最小二乘平差及其在POLInSAR植被高反演中的应用

传统的测量观测值都是实数,因此测量平差都是在实数空间中进行的。然而,随着科学技术的快速发展,现代测绘领域中出现了一些用复数表示的观测数据。与实数数据一样,这些复数数据同样面临着如何从带有误差的观测值中找出未知量的最佳估计值的问题。但目前涉及复数观测的数据处理时,主要还是依据观测过程,分步或直接解算,不能考虑观测误差、多余观测信息等。针对这一情况,本文介绍了复数域中数据处理的最小二乘方法,试图将测量平差从实数域推广到复数域,并定量研究了两种平差准则的优劣性。为了了解复数域最小二乘的有效性,本文以极化干涉SAR植被高反演为例,建立复数域平差函数模型和随机模型,构建复数域最小二乘法反演植被高。结果表明该算法反演的植被高结果可靠,其精度优于经典植被高反演算法,且计算简单,易于实现。     

An iteratively reweighted least-squares approach to adaptive robust adjustment of parameters in linear regression models with autoregressive and <Emphasis Type="Italic">t</Emphasis>-distributed deviations

In this paper, we investigate a linear regression time series model of possibly outlier-afflicted observations and autocorrelated random deviations. This colored noise is represented by a covariance-stationary autoregressive (AR) process, in which the independent error components follow a scaled (Student’s) t-distribution. This error model allows for the stochastic modeling of multiple outliers and for an adaptive robust maximum likelihood (ML) estimation of the unknown regression and AR coefficients, the scale parameter, and the degree of freedom of the t-distribution. This approach is meant to be an extension of known estimators, which tend to focus only on the regression model, or on the AR error model, or on normally distributed errors. For the purpose of ML estimation, we derive an expectation conditional maximization either algorithm, which leads to an easy-to-implement version of iteratively reweighted least squares. The estimation performance of the algorithm is evaluated via Monte Carlo simulations for a Fourier as well as a spline model in connection with AR colored noise models of different orders and with three different sampling distributions generating the white noise components. We apply the algorithm to a vibration dataset recorded by a high-accuracy, single-axis accelerometer, focusing on the evaluation of the estimated AR colored noise model.     

不等式约束加权整体最小二乘的凝聚函数法

误差向量的方差-协方差阵是一般对称正定矩阵下的附不等式约束加权整体最小二乘平差模型,研究了其参数估计和精度评定问题。首先,将残差平方和极小化函数在整体最小二乘准则下转化为只包含模型参数的目标函数,同时将所有的不等式约束表示成一个等价的凝聚约束函数,并运用乘子罚函数策略将不等式约束加权整体最小二乘平差问题转化为相应的无约束最优化问题,并用BFGS方法求解。然后,将误差方程和约束函数线性展开,推导了最优解和观测量间的近似线性函数关系,运用方差-协方差传播律得到了最优解的近似方差。最后,用数值实例验证了方法的有效性和可行性。     

A few basic principles and techniques of array algebra

This paper is intended to demonstrate the usefulness of array algebra techniques in certain multilinear least squares problems. A typical restriction of array algebra is the need for a gridded observational structure; however, the grid does not have to be uniform and in general is not limited to any particular coordinate system nor to two- or three-dimensional spaces. Another restriction comes to light when dealing with weighted multilinear least squares adjustments. The a—priori variance-covariance matrix cannot be completely arbitrary but must be expressible in terms of certain matrix products. There exist various practical ways (not discussed herein) to bridge these restrictions. The reward for using the array algebra technique when it is appropriate lies in the great computational savings. From the theoretical point of view, the backbone of most derivations are the “R-matrix multiplications” and a simple tool, demonstrated herein, called “fundamental transformation”. It follows that the least squares solution of “array observation equations” does not have to be sought by some new and complex mathematical means. The fundamental transformation allows such an adjustment problem to be rewritten in a conventional (monolinear) form; the familiar least squares solution is then written down and transformed back to the array form using the same tool. The statistical properties of the results (e.g. minimum variance) are known from the conventional approach and do not have to be rederived in the array case.     

PEIV模型参数估计新算法

PEIV(Partial Errors-In-Variables)模型是EIV模型的扩展,它能解决系数矩阵含有非随机元素或存在结构特性的问题。针对常规PEIV模型算法的复杂性,提出了一种PEIV模型参数估计的新算法。该算法将系数矩阵含误差的元素看成是一类观测值,与平差模型原观测值构成两类观测值,将PEIV平差模型表示为类似于传统的最小二乘间接平差模型,再通过非线性最小二乘平差理论,推导出了算法的迭代公式和精度评定公式。算法迭代格式与间接平差类似,通过算例验证了算法的可行性和正确性。     

A detailed study of the duality relation for the least squares adjustment in euclidean spaces

The Euclidean spaces with their inner products are used to describe methods of least squares adjustment as orthogonal projections on finite-dimensional subspaces. A unified Euclidean space approach to the least squares adjustment methods “observation equations” and “condition equations” is suggested. Hence not only the two adjustment solutions are treated from the view-point of Euclidean space theory in a unified frame but also the existing duality relation between the methods of “observation equations” and “condition equations” is discussed in full detail. Another purpose of this paper is to contribute to the development of some familiarity with Euclidean and Hilbert space concepts. We are convinced that Euclidean and Hilbert space techniques in least squares adjustment are elegant and powerful geodetic methods.     

半参数回归与模型精化

就一般情况给出了半参数平差的算法，并结合一种特定的情况，讨论了正规化矩阵半正定时的计算方法，给出了相应的公式，最后构造了一个模拟的平差问题，对半参数法和最小二乘法的计算结果进行了比较，计算表明，半参数法能够发现并识别模型误差或观测值中的系统误差。     

实参数与复参数混合域最小二乘平差方法

针对平差问题同时含有实数域参数和复数域参数的情形,提出了实参与复参混合的测量平差建模思路,并导出了相应的混合域最小二乘平差方法。该方法统一概括了实最小二乘和复最小二乘方法,其估计过程包括两步:基于零空间算子的实参数平差估计和复数平差模型的复参数估计。通过等价变换将实数域投影变换模型、直接线性变换模型分别重构为相应的混合域平差模型,有效降低法方程求逆维数,从而提高建模和平差计算效率。应用结果及分析验证了本文所提方法的正确性和有效性。     

Total least squares adjustment in partial errors-in-variables models: algorithm and statistical analysis

The weighted total least squares (TLS) method has been developed to deal with observation equations, which are functions of both unknown parameters of interest and other measured data contaminated with random errors. Such an observation model is well known as an errors-in-variables (EIV) model and almost always solved as a nonlinear equality-constrained adjustment problem. We reformulate it as a nonlinear adjustment model without constraints and further extend it to a partial EIV model, in which not all the elements of the design matrix are random. As a result, the total number of unknowns in the normal equations has been significantly reduced. We derive a set of formulae for algorithmic implementation to numerically estimate the unknown model parameters. Since little statistical results about the TLS estimator in the case of finite samples are available, we investigate the statistical consequences of nonlinearity on the nonlinear TLS estimate, including the first order approximation of accuracy, nonlinear confidence region and bias of the nonlinear TLS estimate, and use the bias-corrected residuals to estimate the variance of unit weight.