Estimation in the presence of many nuisance parameters: Composite likelihood and plug-in likelihood

We consider the incidental parameters problem in this paper, i.e. the estimation for a small number of parameters of interest in the presence of a large number of nuisance parameters. By assuming that the observations are taken from a multiple strictly stationary process, the two estimation methods, namely the maximum composite quasi-likelihood estimation (MCQLE) and the maximum plug-in quasi-likelihood estimation (MPQLE) are considered. For the MCQLE, we profile out nuisance parameters based on lower-dimensional marginal likelihoods, while the MPQLE is based on some initial estimators for nuisance parameters. The asymptotic normality for both the MCQLE and the MPQLE is established under the assumption that the number of nuisance parameters and the number of observations go to infinity together, and both the estimators for the parameters of interest enjoy the standard root-n$n$ convergence rate. Simulation with a spatial–temporal model illustrates the finite sample properties of the two estimation methods.     

An Inverse Problem in Periodic Diffractive Optics: Reconstruction of Lipschitz Grating Profiles

We consider the problem of recovering a two-dimensional periodic structure from scattered waves measured above the structure. Following an approach by Kirsch and Kress, this inverse problem is reformulated as a nonlinear optimization problem. We develop a theoretical basis for the reconstruction method in the case of an arbitrary Lipschitz grating profile. The convergence analysis is based on new perturbation and stability results for the forward problem.     

半变系数模型PLS估计的渐近正态性

半变系数模型在统计建模中具有重要的应用．最近几年，人们提出了许多方法来估计其常系数和函数系数，但是估计的渐近性质还没有被系统的研究．本文介绍了半变系数模型的PLS估计，在Fan和Huang对常系数渐近性质研究的基础上，给出了函数系数的渐近正态性。     

高光谱成像的土壤剖面水分含量反演及制图

传统土壤水分的获取方法仅可获得离散的土壤水分点位数据,难以获得剖面上精细且连续的水分含量分布图。研究了野外条件下利用近红外高光谱（882～1 709 nm）成像反演剖面土壤水分含量（SMC）,并实现精细制图的可行性。研究剖面位于江苏省东台市,我们利用近红外高光谱成像仪对剖面进行了5天原位连续观测,共采集了280个土样用于烘干法测定SMC。原始高光谱图像经数字量化值（DN）校正、黑白校正、拼接、几何校正、剪切和掩膜等一系列预处理后,提取各采样点的平均光谱反射率。提取光谱（Raw）经吸光度[LOG₁₀(1/R)],Savitzky-Golay平滑（SG）、一阶微分（FD）、二阶微分（SD）、多元散射校正（MSC）和标准正态变量（SNV）转换后,采用偏最小二乘回归（PLSR）和最小二乘支持向量机（LS-SVM）方法建立SMC预测模型,并对比分析不同光谱预处理方法与建模方法组合条件下SMC的预测精度。结果表明,光谱反射率随SMC增加逐渐降低,不同光谱预处理方法的预测精度有所差异,除MSC方法外,同一光谱预处理方法的LS-SVM模型预测精度均高于PLSR模型,并且基于LOG₁₀(1/R)光谱的LS-SVM模型对SMC预测精度最高,其建模集的决定系数（R2_c）和均方根误差（RMSE_c）分别为0.96和0.65%,预测集的决定系数（R2_p）、均方根误差（RMSE_p）和相对分析误差（RPD_p）分别为0.88,1.05%和2.88。利用最优模型进行剖面SMC的高空间分辨率精细制图,通过比较SMC反演图中提取的预测值与实测值关系发现预测精度较高（R2: 0.85～0.95, RMSE: 0.94%～1.02%）,且两者在剖面中的变化趋势基本一致,说明SMC反演图不仅能很好地反映出土壤水分在整个剖面中毫米级的含量分布信息,也可反映出同一位置处不同天数间的含量差异。因此,利用近红外高光谱成像结合优化的预测模型,能够实现土壤剖面SMC的定量预测及精细制图,有助于快速、有效监测田间剖面土壤水分状况。     

统一用户数据库的分析和设计

统一用户档案（GUP）是3GPP定义的一种融合用户数据访问框架,它的基本出发点是以用户为中心,解决多种不同网络、业务的用户数据统一访问的问题。本文以3GPP的研究为基础,结合中国运营商网络的特点,对同一用户档案（GUP）在中国运营商网络中的落地进行了分析,提出了统一用户档案数据库（GUPP）的架构、数据定义,并对实现难点进行了剖析。     

剖面不变性与非局部扰动电子温度响应

在ＴＦＴＲ及ＴＥＸＦ等装置上观测到的边界电子冷脉冲的输运现象是通常基于局部热导的瞬态输运理论所无法解释的。利用一个与电子温度剖面不变性相「容的且与在ρ＝ρＬ（ρＬ≈０．７～０．８）处的局部电子热导相关的非局部电子热导模型来描述边界冷脉冲引起的电子温度变化的瞬态过程，发现此模型可定性地解释边界冷脉冲实验现象的几乎所有主要特征。     

部分线性变系数模型的Profile Lagrange乘子检验

对于部分线性变系数模型附有约束条件时的估计与检验问题,基于Profile最小二乘方法给出了参数部分以及非参数部分的约束估计并研究了它们的渐近性质,并针对约束条件构造了Profile Lagrange乘子检验统计量,证明了该统计量在原假设下的渐近分布为χ2分布,从而将Lagrange乘子检验方法推广到了半参数模型上.     

Core temperature and density profile measurements in inertial confinement fusion implosions

We have measured time-integrated and time-gated electron temperature (T_e) and density (N_e) spatial profiles from indirect-drive implosions. In our experiments, we used a multiple-pinhole two-dimensional imaging spectrometer to obtain multispectral X-ray images of the imploded core. Quantitative comparisons between quasi-monochromatic images in different energy bands allowed T_e and N_e spatial profiles to be determined using two independent and validated techniques: a multi-objective search and reconstruction analysis, and an analytical analysis. We then compared the results to a simple one-dimensional (1D) mix-free hydrodynamics simulation in order to evaluate the ability of such a model to predict our experiments. Our data show spatial T_e profiles that are qualitatively consistent with the predictions of our 1D simulations, but we observe central cores that are 10–25% cooler and emit X-rays as late as 200 ps after peak compression. We infer time-gated spatial N_e profiles that are consistent with our 1D simulations near the times of peak compression, but we find significant disagreement between time-integrated data and 1D simulation predictions at large radii. Careful analysis of the time-gated and time-integrated T_e and N_e spatial profiles, together with streaked X-ray emission spectra from core and shell dopants, suggests mixing of shell material into the core is an important process that our 1D hydrodynamics simulations fail to capture, and comparison between image data and a simple analytical model suggests that 2–5 μm of the initial inner shell thickness mixes into the core during the time period of significant X-ray emission. This mix width is consistent with the predictions of a growth-factor analysis that treats instability growth seeded by capsule surface roughness, and points to the need to consider time-dependent mixing effects when interpreting T_e and N_e spatial profiles derived from multispectral X-ray image data, particularly at large radii where mixing effects will be most significant.     

生长材料中掺杂杂质纵向分布的扩散生长理论

从ＭＯＣＶＤ和卤化物ＶＰＥ的生长装置出发，分析了量子阱激光材料和微波电子材料中引入杂质的纵向分布和谐过程。采用了掺入杂质分子经载体气体漂移扩散输运和生长过程中再扩散的数学物理理论，导出了掺杂杂质最终纵向浓度分布的数学定量解析式。根据本理论，提出了陡峭掺杂和均匀纵向浓度分布的工艺解决方案。     

基于联合动态稀疏表示的开集距离像目标识别方法

针对开集条件下多视高分辨距离像(HRRP)目标识别问题,提出了一种基于联合动态稀疏表示(JDSR)的开集识别方法。该方法利用JDSR求解多视HRRP在过完备字典上的重构误差,采用极值理论(EVT)对匹配和非匹配类别的重构误差拖尾进行建模,将开集识别问题转化为假设检验问题求解。识别时利用重构误差确定候选类,根据尾部分布的置信度获得匹配类与非匹配类得分,并将两者的加权和作为类别判据最终确定库外目标或候选类。该方法能够有效利用多视观测来自相同目标的先验信息提高开集条件下的HRRP识别性能,并且对多视数据不同的获取场景具有良好的适应性。利用从MSTAR反演生成的HRRP数据对算法进行了测试,结果表明所提方法的性能优于主流开集识别方法。