Luisa Fermo,Lothar Reichel,Giuseppe Rodriguez,Miodrag M. Spalević
Abstract:Fredholm integral equations of the second kind that are defined on a finite or infinite interval arise in many applications. This paper discusses Nyström methods based on Gauss quadrature rules for the solution of such integral equations. It is important to be able to estimate the error in the computed solution, because this allows the choice of an appropriate number of nodes in the Gauss quadrature rule used. This paper explores the application of averaged and weighted averaged Gauss quadrature rules for this purpose, and introduces new stability properties for them.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the solution of Fredholm integral equations of the second kind on finite or infinite intervals. Specifically, the paper explores the Nyström method based on Gauss quadrature rules to solve this type of integral equations, and focuses on how to estimate the error in the computed solution. This includes the introduction of the application of the averaged and weighted - averaged Gauss quadrature rules, as well as the new stability properties of these rules, in order to more effectively select the appropriate number of nodes of the Gauss quadrature rules, thereby improving the accuracy and reliability of the solution.
### Background and Problem Description of the Paper
The Fredholm integral equation of the second kind has the following form:
\[ f(y)+\int_D k(x, y)f(x) \, d\mu(x)=g(y), \quad y \in D, \]
where the kernel function \( k \) and the right - hand - side function \( g \) are known, and the function \( f \) needs to be determined, while \( d\mu(x) \) is a non - negative measure defined on a bounded or unbounded domain \( D \). This type of equations appears in many applications, such as image restoration (using Tikhonov regularization), conformal mapping, frequency analysis, and tomography, etc.
### Nyström Method
The Nyström method is a commonly used numerical method for solving Fredholm integral equations of the second kind. This method transforms the integral equation into a linear system of equations by replacing the integral with an interpolation quadrature rule. The specific steps are as follows:
1. **Integral Approximation**: Replace the integral with the Gauss quadrature rule \( K_m \) with \( m \) nodes \( x_1, x_2, \ldots, x_m \):
\[ (I + K_m)f_m = g, \]
where \( I \) is the identity operator and \( f_m \) is the unknown interpolation function.
2. **Linear System of Equations**: Require the above equation to hold at the nodes \( y = x_i \), and obtain a linear system of equations:
\[ \sum_{j = 1}^m[\delta_{ij}+c_j k(x_j, x_i)]a_j = g(x_i), \quad i = 1, 2, \ldots, m, \]
where \( \delta_{ij} \) is the Kronecker delta function and \( a_j = f_m(x_j) \).
3. **Solution Computation**: After solving the linear system of equations, obtain the coefficients \( a_1, a_2, \ldots, a_m \), and then construct the Nyström interpolation function:
\[ f_m(y)=g(y)-\sum_{j = 1}^m c_j k(x_j, y)a_j, \quad y \in D. \]
### Error Estimation and Improvement
An important aspect of the paper is the selection of appropriate quadrature formulas, especially the Gauss quadrature rules. In order to estimate the error of the Nyström interpolation function, the paper introduces the averaged and weighted - averaged Gauss quadrature rules. These rules provide higher accuracy and stability by combining two Gauss quadrature rules with different numbers of nodes.
### Main Contributions
1. **New Stability Results**: The paper shows the stability and convergence of the averaged and weighted - averaged Gauss quadrature rules in the weighted function space.
2. **Error Estimation**: Proposes a method for estimating the error of the Nyström interpolation function using the averaged and weighted - averaged rules.
3. **Iterative Methods**: Develops new iterative methods for solving the linear system of equations related to the Nyström method. These methods utilize the structure of the coefficient matrix and improve the computational efficiency.
### Conclusion
By introducing and analyzing the averaged and weighted - averaged Gauss quadrature rules, the paper provides a more accurate and reliable method for the numerical solution of Fredholm integral equations of the second kind. These methods not only improve the accuracy of the solution but also reduce the amount of computation, and have important theoretical and practical application values.