A trigonometrical approach for some projection methods
In order to obtain an approximate solution of a Fredholm integral equation of the first kind \(Kf=g,\;(Kf)(s) = \int_a^b k(s, t)f(t) dt,\;k\in L_2([a, b] \times [a, b]),\;K: L_2[a, b]\to L_2[a, b],\) an abstract projection method is proposed. Its convergence is characterized in terms of the minimal angle between the range and nullspace of the projections, and expressed as a compatibility condition between the approximation spaces and the operator \(K.\) It is optimal for approximation spaces generated by singular functions of \(K.\) Convergence of some known projection methods is analyzed as particular cases of the general method. Numerical methods for calculating the angles, based on eigenvalues of pencils of matrices, are given in each special case. As an example, the ``\(p\)-iterated integration operator \((Kf)(s) = \int_0^s (s- t)^pf(t) dt,\;K: L_2[0, 1]\to L_2[0, 1],\) is considered.
- A modification of the projection method for integral equation of the first kind
- Projected Tikhonov regularization method for Fredholm integral equations of the first kind
- scientific article; zbMATH DE number 6020658 (Why is no real title available?)
- A Variant of Projection-Regularization Method for Ill-Posed Linear Operator Equations
- Extensions of projection and collocation methods for first kind integral equations
- scientific article; zbMATH DE number 3922721 (Why is no real title available?)
- Mean value projection method of Fredholm integral operator
- scientific article; zbMATH DE number 6258186 (Why is no real title available?)
This page was built for publication: A trigonometrical approach for some projection methods
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1300265)