A New Preconditioning Technique of the LSQR Method for Solving the Linear Systems

The LSQR method is an attractive method for solving the linear systems. However,LSQR may converge slowly or even stagnate due to a limited Krylov subspace. To ameliorate this drawback, a polynomial preconditioned variant of LSQR is presented.Numerical experiments from real applications are used to validate the e ectiveness ofthe proposed method.

کلیدواژه ها:

LSQR method ، Krylov subspace methods ، Preconditioned LSQR.