The Levels-Recursive Algorithm for Vector Valued Interpolants by Triple Branched Continued Fractions

1 Introduction Let Πl,m,n be a set of points in three dimensional space R3, Πl,m,n = {(xi, yj, zk), i = 0, 1, · · · l; j = 0, 1, · · · m; k = 0, 1, · · · n}. Let a d?dimensional vector vi,j,k be given at every point (xi, yj, zk) ∈ Πl,m,n and     

连续时间下局部光滑统计量的a.s.收敛速度和边界效应

本文在连续时间场合研究回归函数的非参数估计量之一———局部光滑统计量的性质 .不仅给出其a.s.收敛的一个速度 ,而且证明了该统计量不受边界效应影响的优良性质 ;同时指出了在连续时间的场合 ,估计量中窗宽的选择和样本的轨道性质都对该估计量的性质产生重要的影响     

（h，φ）-不变广义凸函数的若干性质与（h，φ）-不变广义凸多目标规划的最优性及对偶性

研究了Ben-Tal广义代数运算的若干性质，引进了(h，φ)-不变广义凸函数的概念，讨论了(h，φ)-不变广义凸函数的若干性质，给出了(h，φ)-不变广义凸半无限多目标规划取得有效解(或弱有效解)的充分条件，建立了(h，φ)-不变广义凸多目标规划的Lagrange对偶理论。     

三参数Weibull分布定数截尾下形状参数的区间估计

本给出了在定数截尾场合下利用三个次序统计量来构造三参数Weibull分布形状参数精确置信区间的方法，另外本还给出了三参数Weibull分布三个参数的联合置信域。章的最后，给出了Monte-Carlo模拟例子并运用本的方法求解了两个实例。     

超线性与二次收敛序列线性方程组算法

本文,在无严格互补条件下,对非线性不等式约束最优化问题提出了一个新的序列线性方程组(简称SSLE)算法．算法有两个重要特征:首先,每次迭代,只须求解一个线性方程组或一个广义梯度投影阵,且线性方程组可以无解．其次,初始点可以任意选取．在无严格互补条件下,算法仍有全局收敛性、强收敛性、超线性收敛性及二次收敛性．文章的最后,还对算法进行了初步的数值实验．     

Experimental studies on the multistage constant modulus array for the estimation of directions-of-arrival

The performance of the multistage Constant Modulus (CM) array, one of the most striking blind beamforming algorithms, for the source Directions-of-Arrival (DOA) estimation was analyzed via computer simulations and water tank experiments, and was compared to that of other DOA estimation algorithms including the ‘non-blind‘ and the ‘blind‘. Firstly, a nominal array model and an array model with gain and phase perturbations were established,respectively. Secondly, the multistage CM array algorithm was described and computer simulations were conducted. Simulation results showed that the multistage CM array could correctly estimate the DOA at the same time when the sources were blindly recovered, and the angle separating ability of the algorithm was beyond the Rayleigh resolution limit. By changing the variance of the array model errors, it was also verified that the multistage CM array was more robust to the errors than some other algorithms mentioned in this paper. Finally, water tank experiments and data processing results were provided. Situations with different array sizes and source angular separations were considered. It was shown that the results were in good agreement with computer simulations. Results of computer simulations and water tank experiments verified that the DOAs of the multiple independent sources could be blindly and robustly estimated with the multistage CM array.     

钚体源样品γ能谱计算的蒙特卡罗方法

提出了钚体源样品γ能谱分步计算的蒙特卡罗模型和方法.采用分步和直接计算两种方法对球壳形钚体源样品的γ能谱进行了模拟计算.用实验测量的γ能谱验证了计算结果的可靠性，计算结果与实验测量值在10%的误差范围符合.通过对Steve Fetter模型的模拟计算表明：对于外面包有若干屏蔽层的钚体源样品大系统，采用分步方法进行能谱计算，可以使结果快速收敛.     

环Z/p~kZ上的两类矩阵方程

利用环上矩阵方法,研究了环Z/pkZ上的两类矩阵方程,确定了这两类矩阵方程的解数.     

协方差改进估计中的变量选择

本文探讨了如何选择协变量的问题, 提出了协变量选择的典则相关检验法, 给出了检验统计量的近似分布, 并且在经济、生物和医药等方面常见的三种误差协方差阵的假定下进行了计算机模拟, 其结果显示了协变量选择的必要性及典则相关检验法的优良性.     

直觉Menger空间中的广义压缩映射原理及其在微分方程中的应用

利用Atanassov的思路,将直觉Menger空间定义为由Menger提出的Menger空间的自然推广.同时也得出一个新广义压缩映射,并运用该压缩映射证明了直觉Menger空间中微分方程解的存在性定理.