Stability of T. Chan＇s Preconditioner from Numerical Range

A matrix is said to be stable if the real parts of all the eigenvalues are negative. In this paper, for any matrix An, we discuss the stability properties of T. Chan＇s preconditioner cU（An） from the viewpoint of the numerical range. An application in numerical ODEs is also given.     

ALTERNATELY LINEARIZED IMPLICIT ITERATION METHODS FOR SOLVING QUADRATIC MATRIX EQUATIONS

A numerical solution of the quadratic matrix equations associated with a nonsingular M-matrix by using the alternately linearized implicit iteration method is considered. An iteration method for computing a nonsingular M-matrix solution of the quadratic matrix equations is developed, and its corresponding theory is given. Some numerical examples are provided to show the efficiency of the new method.     

MODIFIED BERNOULLI ITERATION METHODS FOR QUADRATIC MATRIX EQUATION

We construct a modified Bernoulli iteration method for solving the quadratic matrix equation AX^2 ＋ BX ＋ C = 0, where A, B and C are square matrices. This method is motivated from the Gauss-Seidel iteration for solving linear systems and the ShermanMorrison-Woodbury formula for updating matrices. Under suitable conditions, we prove the local linear convergence of the new method. An algorithm is presented to find the solution of the quadratic matrix equation and some numerical results are given to show the feasibility and the effectiveness of the algorithm. In addition, we also describe and analyze the block version of the modified Bernoulli iteration method.     

用径向函数法求解偏微分方程

In this paper. Kansa′s method and Hermite collocation method with Radial Basis Func-tions is applied to solve partial differential equation. The resultant matrix generated from the Her-mite method is positive definite, which guarantees the reversibility of the matrix. The numerical re-sults indicate that the methods provides reversibility of the matrix. The numerical results indicatethat the method provieds an efficient algorithm for solving partial differential equations.     

MULTI-SCALE METHODS FOR INVERSE MODELING IN 1-D MOS CAPACITOR

In this paper,we investigate multi-scale methods for the inverse modeling in 1-D Metal-Oxide-Silicon(MOS) capactior,First,the mathematical model of the device is given and the numerical simulation for the forward problem of the model is implemented using finite element method with adaptive moving mesh. Then numerical analysis of these parameters in the model for the inverse problems is presented .Some matrix analysis tools are applied to explore the parameters‘ sensitivities,And thired,the parameters are extracted using Levenberg-Marquardt optimization method.The essential difficulty arises from the effect of multi-scale physical differeence of the parameters.We explore the relationship between the parameters‘ sensitivitites and the sequencs for optimization,which can seriously affect the final inverse modeling results.An optimal sequence can efficiently overcome the multip-scale problem of these parameters,Numerical experiments show the efficiency of the proposed methods.     

The coninvolutory decomposition and its computation for a complex matrix

A complex, square matrix E is called coninvolutory if EE = I, where E denotes complex conjugate of the matrix E and I is an identity matrix. In this paper we introduce the coninvolutory decomposition of a complex matrix and investigate a Newton iteration for computing the coninvolutory factor. A simple numerical example illustrates our results.     

ERROR ANALYSIS FOR A FAST NUMERICAL METHOD TO A BOUNDARY INTEGRAL EQUATION OF THE FIRST KIND

For two-dimensional boundary integral equations of the first kind with logarithmic kernels, the use of the conventional boundary element methods gives linear systems with dense matrix. In a recent work [J. Comput. Math., 22 （2004）, pp. 287-298], it is demonstrated that the dense matrix can be replaced by a sparse one if appropriate graded meshes are used in the quadrature rules. The numerical experiments also indicate that the proposed numerical methods require less computational time than the conventional ones while the formal rate of convergence can be preserved. The purpose of this work is to establish a stability and convergence theory for this fast numerical method. The stability analysis depends on a decomposition of the coefficient matrix for the collocation equation. The formal orders of convergence observed in the numerical experiments are proved rigorously.     

$D^2(\Omega,\rho)$的再生核与再生核诱导的度量

An important property of the reproducing kernel of D^2（Ω, ρ） is obtained and the reproducing kernels for D^2（Ω, ρ） are calculated when Ω = Bn× Bn and ρ are some special functions. A reproducing kernel is used to construct a semi-positive definite matrix and a distance function defined on Ω×Ω. An inequality is obtained about the distance function and the pseudodistance induced by the matrix.     

RESEARCH ANNOUNCEMENTS—On a Generalization of the c—numerical Range

Let Cn×n be the set of n × n complex matrices and An the set of orthonormal n-tuples of vectors in Cn. For a vector c in Cn and a matrix A in Cn×n, the c-numerical range of A is the set Wc(A)={n∑i=1 Ci(Axi,xi):(x1,…xn)∈∧n} When c = (1,0,…,0), Wc(A) is reduced to the classical numerical range W(A) (see [1]). For the classical numerical range and its generalizations, one may see the survey article[2].     

Least-Squares Solution with the Minimum-Norm for the Matrix Equation A^TXB ＋ B^TX^TA = D and Its Applications

An efficient method based on the projection theorem,the generalized singular value decompositionand the canonical correlation decomposition is presented to find the least-squares solution with the minimum-norm for the matrix equation A~TXB B~TX~TA=D.Analytical solution to the matrix equation is also derived.Furthermore,we apply this result to determine the least-squares symmetric and sub-antisymmetric solution ofthe matrix equation C~TXC=D with minimum-norm.Finally,some numerical results are reported to supportthe theories established in this paper.     

Numerical scheme to solve a class of variable–order Hilfer–Prabhakar fractional differential equations with Jacobi wavelets polynomials

In this paper,we introduced a numerical approach for solving the fractional differential equations with a type of variable-order Hilfer-Prabhakar derivative of orderμ(t)andν(t).The proposed method is based on the Jacobi wavelet collocation method.According to this method,an operational matrix is constructed.We use this operational matrix of the fractional derivative of variable-order to reduce the solution of the linear fractional equations to the system of algebraic equations.Theoretical considerations are discussed.Finally,some numerical examples are presented to demonstrate the accuracy of the proposed method.     

AN APPROACH TO DESIGN OBSERVER FOR A CLASS OF DISCRETE CONTROL SYSTEMS WITH TIME-DELAY

The design of functional observer for a class of discrete systems with time-delay is concerned. The solution to Sylvester function is given. An improved method for the functional observer design is proposed by the condition of linear matrix inequality. In the end,an example is given to illustrate the feasibility of the method.     

MINIMIZATION PROBLEM FOR SYMMETRIC ORTHOGONAL ANTI-SYMMETRIC MATRICES

By applying the generalized singular value decomposition and the canonical correlation decomposition simultaneously, we derive an analytical expression of the optimal approximate solution ^-X, which is both a least-squares symmetric orthogonal anti-symmetric solu- tion of the matrix equation A^TXA = B and a best approximation to a given matrix X^＊. Moreover, a numerical algorithm for finding this optimal approximate solution is described in detail, and a numerical example is presented to show the validity of our algorithm.     

A Fast Numerical Method for Integral Equations of the First Kind with Logarithmic Kernel Using Mesh Grading

The aim of this paper is to develop a fast numerical method for two-dimensional boundary integral equations of the first kind with logarithm kernels when the boundary of the domain is smooth and closed. In this case, the use of the conventional boundary element methods gives linear systems with dense matrix. In this paper, we demonstrate that the dense matrix can be replaced by a sparse one if appropriate graded meshes are used in the quadrature rules. It will be demonstrated that this technique can increase the numerical efficiency significantly.     

离散神经网络指数稳定的增强李亚普诺夫方法

This paper addresses the problem of robust stability for a class of discrete-time neural networks with time-varying delay and parameter uncertainties.By constructing a new augmented Lyapunov-Krasovskii function,some new improved stability criteria are obtained in forms of linear matrix inequality（LMI） technique.Compared with some recent results in the literature,the conservatism of these new criteria is reduced notably.Two numerical examples are provided to demonstrate the less conservatism and effectiveness of the proposed results.     

反中心对称矩阵的广义特征值反问题

Given matrix X and diagonal matrix A , the anti-centrosymmetric solutions (A, B) and its optimal approximation of inverse generalized eigenvalue problem AX = BXA have been considered. The general form of such solutions is given and the expression of the optimal approximation solution to a given matrix is derived. The algorithm and one numerical example for solving optimal approximation solution are included.     

A type of new posteriori error estimators for stokes problems

1. IntroductionIn the numerical approximation of PDE, it is often very importals to detect regionswhere the accuracy of the numerical solution is degraded by local singularities of the solutionof the continuous problem such as the singularity near the re-entrant corller. An obviousremedy is to refine the discretization in the critical regions, i.e., to place more gridpointswhere the solution is less regular. The question is how to identify these regions automdticallyand how to determine a goo…     

A new algorithm for pole assignment of single-input linear systems using state feedback

In this paper we present a new algorithm for the single-input pole assignment problem using state feedback. This algorithm is based on the Schur decomposition of the closed-loop system matrix, and the numerically stable unitary transformations are used whenever possible, and hence it is numerically reliable.The good numerical behavior of this algorithm is also illustrated by numerical examples.     

C~2 CONTINUOUS NEARLY ARC-LENGTH PARAMETERIZED CURVE

An efficient method for C~2 nearly arc-length parameterized curve is presented. An idea of approximation for the arc-length function of parametric curve which interpolates CAD data points is discussed. The parameterization is implemented by using parameter transformation. Finally, two numerical examples are given..     

QR factorization for row or column symmetric matrix

The problem of fast computing the QR factorization of row or column symmetric matrix is considered. We address two new algorithms based on a correspondence of Q and R matrices between the row or column symmetric matrix and its mother matrix. Theoretical analysis and numerical evidence show that, for a class of row or column symmetric matrices, the QR factorization using the mother matrix rather than the row or column symmetric matrix per se can save dramatically the CPU time and memory without loss of any numerical precision.