scientific article; zbMATH DE number 3850444
[For the entire collection see Zbl 0499.00015.] Discretization of an integral equation eigenvalue problem \(Kx=\lambda x\) by Galerkin's method produces a generalized matrix eigenvalue problem \(C_ h\xi_ h=\lambda_ hG_ h\xi_ h\) where \(C_ h\) is the stiffness matrix and \(G_ h\) is the Gramian matrix of the trial functions. The matrix eigenvalue problem can be brought into a symmetric form by Cholesky decomposition of the Gramian \(G_ h\), \(G_ h=L_ h\times L_ h\!^ t\). The only essential drawback of symmetrization by Cholesky decomposition is the repeated square rooting. The subject of this papeer is the construction of special trial functions with the property that the Cholesky decomposition can be done in advance, once and for all.
- On trial functions applied in the generalized Trefftz method
- Fast algorithms using orthogonal polynomials
- Fast Fourier orthogonalization
- Fast Fourier Orthogonalization over NTRU Lattices
- A fast algorithm for evaluation of normalized Hermite functions
- Orthogonal compactly supported functions and numerical methods
- scientific article; zbMATH DE number 4137264
- Fast conversion algorithms for orthogonal polynomials
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3319745)