scientific article; zbMATH DE number 17457
[For the entire collection see Zbl 0728.00018.] During the last decade, the authors have published several articles dealing with the efficiency of the conjugate gradient (CG) algorithm for solution of extreme eigenproblems of symmetric matrices. An advantage of this algorithm is the fact that it finds the minimum eigenvector directly, while the conventional methods, such as the QR algorithm, solve, needed or not, all the eigenvalues and eigenvectors simultaneously. It turns out that the introduction of the fast Fourier transform (FFT) into the computation further improves the efficiency of the QR algorithm and demonstrates its flexibility. In the present paper these ideas are applied to a minimum eigenproblem of a Hankel system. Such systems arise, as the authors state, in many areas of digital signal processing. The authors solve a numerical example both in terms of the CG algorithm and in terms of the EIGRS routine based on the QR algorithm. The result given by the former method and using 5 ms. CPU time is more accurate than that given by the latter method and using 27 ms. CPU time.
- Efficient solution of a Toeplitz-plus-Hankel coefficient matrix system of equations
- CONJUGATE GRADIENT METHODS FOR SOLVING THE SMALLEST EIGENPAIR OF LARGE SYMMETRIC EIGENVALUE PROBLEMS
- Computing the smallest eigenvalue of large ill-conditioned Hankel matrices
- Conjugate gradient methods for the Rayleigh quotient minimization of generalized eigenvalue problems
- A fast eigenvalue algorithm for Hankel matrices
- An efficient algorithm on optimal Hankel-norm approximation for multivariable systems
- Parallel and superfast algorithms for Hankel systems of equations
- A frequency condensation method for the eigenvalue problem
- A generalized conjugate gradient method for eigenvalue problems
- Computing several eigenpairs of Hermitian problems by conjugate gradient iterations
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 Q3974777)