Deconvolving compactly supported densities

This paper addresses the statistical problem of density deconvolution under the condition that the density to be estimated has compact support. We introduce a new estimation procedure, which establishes faster rates of convergence for smooth densities as compared to the optimal rates for smooth densities with unbounded support. This framework also allows us to relax the usual condition of known error density with non-vanishing Fourier transform, so that a nonparametric class of densities is valid; therefore, even the shape of the noise density need not be assumed. These results can also be generalized for fast decaying densities with unbounded support. We prove optimality of the rates in the underlying experiment and study the practical performance of our estimator by numerical simulations.      

Refractivity estimations from an angle-of-arrival spectrum

This paper addresses the probability of atmospheric refractivity estimation by using field measurements at an array of radio receivers in terms of angle-of-arrival spectrum.Angle-of-arrival spectrum information is simulated by the ray optics model and refractivity is expressed in the presence of an ideal tri-linear profile.The estimation of the refractivity is organized as an optimization problem and a genetic algorithm is used to search for the optimal solution from various trial refractivity profiles.Theoretical analysis demonstrates the feasibility of this method to retrieve the refractivity parameters.Simulation results indicate that this approach has a fair anti-noise ability and its accuracy performance is mainly dependent on the antenna aperture size and its positions.     

Estimation of coefficient of variation in a weighted inverse Gaussian model

The coefficient of variation is an important parameter in many physical, biological and medical sciences. In this paper we study the estimation of the square of the coefficient of variation in a weighted inverse Gaussian model which is a mixture of the inverse Gaussian and the length biased inverse Gaussian distribution. This represents a rich family of distributions for different values of the mixing parameter and can be used for modelling various life testing situations. The maximum likelihood as well as the Bayes estimates of the parameters are obtained. These estimates are used to derive the estimates of the square of the coefficient of variation of the model under study. Several important data sets are analysed to illustrate the results. © 1996 John Wiley & Sons, Ltd.     

单幅条纹图相位解调的小波分析方法

相位解调是条纹相位分析的关键问题.本文提出一种应用小波频率估计联合频率导数对变形条纹进行瞬时频率分析,从中提取参考基频,从而依靠单一变形条纹实现相位解调的方法.首先,理论上证明了当变形条纹瞬时频率空间导数等于零,该空间点的瞬时频率等于参考基频频率;其次,引入Gabor小波提取变形条纹的瞬时频率空间分布,利用变形条纹瞬时频率的空间导数分布识别提取参考基频,从而实现相位解调.利用该方法进行了三维形貌测量的实验,结果表明该方法在实现相位解调中效果良好.     

一点到一仿射集合投影的加权扰动分析

1引言设AEC“”,M和N分别为m和,;阶Hermite正定阵,则存在唯一的矩阵X6C”””满足AXA一A,XAX一X,（MAX）”一MAX,（NXA）”一NXA,（．1）其中B”为B的共轭转置阵．称满足（互．l）的X为A的加权M－P逆,记作X二A和．特别,当M—I。,N—I。时,A;一A”为矩阵的M－P逆．文献［3—5,11,13］讨论了一点到一仿射集合投影的挑动和加权扰动理论．文献［7,9,18」研究了极小N一范数、M一最小二乘解的扰动分析．本文推广了以上文献的主要结果．考虑如下问题「1,2〕：给定AE〔叩“”,bEC一和广EO,找一个向量X”E…     

Four-dimensional parameter estimation of plane waves using swarming intelligence

This paper proposes an efficient approach for four-dimensional(4D) parameter estimation of plane waves impinging on a 2-L shape array. The 4D parameters include amplitude, frequency and the two-dimensional(2D) direction of arrival,namely, azimuth and elevation angles. The proposed approach is based on memetic computation, in which the global optimizer, particle swarm optimization is hybridized with a rapid local search technique, pattern search. For this purpose,a new multi-objective fitness function is used. This fitness function is the combination of mean square error and the correlation between the normalized desired and estimated vectors. The proposed hybrid scheme is not only compared with individual performances of particle swarm optimization and pattern search, but also with the performance of the hybrid genetic algorithm and that of the traditional approach. A large number of Monte–Carlo simulations are carried out to validate the performance of the proposed scheme. It gives promising results in terms of estimation accuracy, convergence rate, proximity effect and robustness against noise.     

部分线性模型中估计的收敛速度

考虑回归模型（Ⅰ）：其中（ｘ＿ｉ，ｔ＿ｉ）是固定非随机设计点列，ｘ＿ｉ＝（ｘ＿（ｉｌ），…，ｘ＿（ｉｐ））＇β＝（β＿１，…，β＿ｐ）＇（ｐ＞１），ｇ是定义在［０，１］上的未知函数，β是未知待估参数，０＜ｔ＿ｉ＜１，ｅ＿ｉ是ｉ．ｉ．ｄ．随机误差，且Ｅｅ＿ｉ＝０，Ｅｅ＝σ￣２＜∞。基于ｇ的估计取一类非参数权估计（包括常见的核估计和近邻估计），我们讨论了β的最小二乘估计及ｇ的估计的最优强弱收敛速度。     

分布形式未知的矩法<英>

我们考虑分布形式未知的矩法,假设分布函数F_θ(x)可用一多项式表示或逼近。多项式的阶数r未知,其系数为未知参数θ_1,…,θ_r。我们给出了θ_1,…,θ_r的一种新的矩法估计,r可自动地求得,估计是强一致的。     

Nonlinear wavelet estimation of regression function with random desigm

The nonlinear wavelet estimator of regression function with random design is constructed. The optimal uniform convergence rate of the estimator in a ball of Besov spaceB ³ _p,q is proved under quite general assumpations. The adaptive nonlinear wavelet estimator with near-optimal convergence rate in a wide range of smoothness function classes is also constructed. The properties of the nonlinear wavelet estimator given for random design regression and only with bounded third order moment of the error can be compared with those of nonlinear wavelet estimator given in literature for equal-spaced fixed design regression with i.i.d. Gauss error. Project supported by Doctoral Programme Foundation, the National Natural Science Foundation of China (Grant No. 19871003) and Natural Science Fundation of Heilongjiang Province, China.     

鲁棒的车载红外视频电子稳像算法

针对车辆行进中,红外热像仪拍摄的视频序列存在复杂的随机抖动,提出基于BRISK特征点匹配的运动估计算法,计算出高精度全局运动矢量,同时对于特征点匹配时出现误匹配及场景中存在前景运动物体的情况,采用模糊聚类法分离全局运动和局部运动,提高了算法的鲁棒性。提出了基于Kalman粒子滤波算法,有效实现了复杂扫描运动和随机抖动的分离,并利用双线性插值法进行图像补偿。采用快速图像拼接法进行未定义区域处理,实现了图像全景输出。还利用车载红外热像仪实际拍摄的红外视频进行了稳像实验。实验结果表明,视频序列获得了很好的稳像效果,能够满足实际应用要求。