
doi: 10.1137/0729066
The concept of a model operator as a preconditioner is one of the widely used for the solution of large elliptic grid systems. During the last decade it gained a recognition in some ``nonelliptic'' applications like the solution of systems with Toeplitz matrix \(A\) which arise, for example, in signal processing and control theory. The authors present a very interesting theory dealing with circulant preconditioners \(C=C^*>0\). They suggest to obtain the known cases of \(C\) on the base of several convolutions of the generating function of \(A\) with specified kernels. The case of the Dirac delta function leads to a new \(C\). All constructions give spectra of \(C^{-1}A\) clustered around 1, thus, the rapid convergence of the modified (preconditioned) conjugate gradient method is predicted.
circulant matrix, Iterative numerical methods for linear systems, convergence, generating function, conjugate gradient method, Numerical computation of matrix norms, conditioning, scaling, circulant preconditioners, Toeplitz matrix, kernels
circulant matrix, Iterative numerical methods for linear systems, convergence, generating function, conjugate gradient method, Numerical computation of matrix norms, conditioning, scaling, circulant preconditioners, Toeplitz matrix, kernels
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 42 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Top 10% | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Top 10% | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Top 10% |
