On the Gaussian Integration of Chebyshev Polynomials

نویسنده

  • P. Rabinowitz
چکیده

It is shown that as m tends to infinity, the error in the integration of the Chebyshev polynomial of the first kind, T{im+2)j±2^x), by an /n-point Gauss integration rule approaches (-!> • 2/(4/2 1), / = 0, 1, ■ • • , m 1, and (-!>' • tt/2, / = m, for all J. 1. Knowledge of the errors in the numerical integration of Chebyshev polynomials of the first kind, Tn(x), by given integration rules has proved to be useful in various situations. On the one hand, they can be used in estimating the error in integrating functions of low-order continuity [4] or with branch-point singularities [5]. On the other hand, they are needed in computing the norm of the error functional of the given rule in a certain family of Hilbert spaces of analytic functions [7]. For certain rules, namely for Gauss, Lobatto and Radau rules, asymptotic, and in some cases exact, values of these errors for certain values of the parameters involved were given in [3]. In this paper, we give further asymptotic results for the case of Gauss integration which are valid for all values of the parameters, thus completing the picture in this particular but very important case. These asymptotic results, which agree with the true computed results quite early, have been used to explain why the use of a Gauss rule with an even number of points, say In, is superior to one with an odd number of points, 2n + 1, in integrating a function which is not analytic at the midpoint of the integration interval [6]. Numerical evidence suggests that similar results hold for a second family of important rules, the Lobatto rules, but the tools at our disposal in this case are not as powerful as in the well-investigated Gauss case and, hence, we were unable to prove these results. 2. The Chebyshev polynomials of the first kind, Tn(x), are defined in I = [— 1, 1] by Tn(x) = cos nd, where x = cos 0, 0 ;£ 0 — it. We shall denote the error in integrating T„(x), using a Gauss m-point rule, by /l m Tn(x)dx £ N>,ltt*) •1 i-l where xt = xi-m, i = 1, • • • , m, are the zeros of the Legendre polynomial of degree m and Wt = witm, i = I, • ■ • , m, are the corresponding weights which are all positive with E"-i w< = 2. Since Tn(x) is odd for n odd and since the Gauss rules are symmetric about the origin, it follows that Em(T2k+1) = 0 for all k. Furthermore, an m-point Gauss rule is exact for all polynomials of degree :S2m — 1. Hence, we shall only consider Em(T2k), where k m. In this case, we have Received March 12, 1971, revised June 6, 1971. AMS 1970 subject classifications. Primary 65D30; Secondary 33A65.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Application of Chebyshev Polynomials for Solving Abel's Integral Equations of the First and Second Kind

In this paper, a numerical implementation of an expansion method is developed for solving Abel's integral equations of the first and second kind. The solution of such equations may demonstrate a singular behaviour in the neighbourhood of the initial point of the interval ofintegration. The suggested method is based on the use of Taylor series expansion to overcome the singularity which le...

متن کامل

A spectral method based on the second kind Chebyshev polynomials for solving a class of fractional optimal control problems

In this paper, we consider the second-kind Chebyshev polynomials (SKCPs) for the numerical solution of the fractional optimal control problems (FOCPs). Firstly, an introduction of the fractional calculus and properties of the shifted SKCPs are given and then operational matrix of fractional integration is introduced. Next, these properties are used together with the Legendre-Gauss quadrature fo...

متن کامل

A numerical Algorithm Based on Chebyshev Polynomials for Solving some Inverse Source Problems

In this paper‎, two inverse problems of determining an unknown source term in a parabolic‎ equation are considered‎. ‎First‎, ‎the unknown source term is ‎estimated in the form of a combination of Chebyshev functions‎. ‎Then‎, ‎a numerical algorithm based on Chebyshev polynomials is presented for obtaining the solution of the problem‎. ‎For solving the problem‎, ‎the operational matrices of int...

متن کامل

On Group Fourier Analysis and Symmetry Preserving Discretizations of PDEs

In this paper we review some group theoretic techniques applied to discretizations of PDEs. Inspired by the recent years active research in Lie groupand exponential time integrators for differential equations, we will in the first part of the article present algorithms for computing matrix exponentials based on Fourier transforms on finite groups. As an example, we consider spherically symmetri...

متن کامل

Stable Gaussian radial basis function method for solving Helmholtz equations

‎Radial basis functions (RBFs) are a powerful tool for approximating the solution of high-dimensional problems‎. ‎They are often referred to as a meshfree method and can be spectrally accurate‎. ‎In this paper, we analyze a new stable method for evaluating Gaussian radial basis function interpolants based on the eigenfunction expansion‎. ‎We develop our approach in two-dimensional spaces for so...

متن کامل

A fractional type of the Chebyshev polynomials for approximation of solution of linear fractional differential equations

In this paper we introduce a type of fractional-order polynomials based on the classical Chebyshev polynomials of the second kind (FCSs). Also we construct the operational matrix of fractional derivative of order $ gamma $ in the Caputo for FCSs and show that this matrix with the Tau method are utilized to reduce the solution of some fractional-order differential equations.

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2010