A MATRIX REDUCTION BASED ALGORITHM TO SOLVE k-ALMOST NORMAL SYSTEMS
Abstract
In this paper, we develop an efficient algorithm to
solve linear system Ax = b where the coefficient matrix A is kalmost normal. We propose an algorithm based on the orthogonalization of the Kyrylov subspace and reduction of the k-almost
normal matrix A to a block tridiagonal form. A comparison with
the popular GMRES method shows that the proposed algorithm
is efficient and in many particular cases generates more accurate
results.
solve linear system Ax = b where the coefficient matrix A is kalmost normal. We propose an algorithm based on the orthogonalization of the Kyrylov subspace and reduction of the k-almost
normal matrix A to a block tridiagonal form. A comparison with
the popular GMRES method shows that the proposed algorithm
is efficient and in many particular cases generates more accurate
results.
Refbacks
- There are currently no refbacks.