symmetric component-wise perturbation
- From: paya <pavel.jiranek@xxxxxxxxx>
- Date: Sun, 12 Aug 2007 16:35:54 -0700
Hello.
I'm trying solve the following problem and I didn't find any reference
to the solution so far.
I'm solving a system of linear equations Ax = b with a sparse matrix A
which is symmetric and positive definite. If I have an approximate
solution y to the exact one x with the residual r = b-Ay, it is known
how to compute the perturbation matrix E with the same structure as A
such that (A+E)y=b, i.e., such that y is the solution of some
perturbed problem and the perturbation is somehow restricted.
I would be very happy if someone could be able to compute the E such
that it is also symmetric or give me a reference to some paper where
this is solved.
Thank you in advance
Pavel J.
.
- Follow-Ups:
- Re: symmetric component-wise perturbation
- From: Carl Barron
- Re: symmetric component-wise perturbation
- Prev by Date: implementing rayleighchan in OFDM
- Next by Date: Re: symmetric component-wise perturbation
- Previous by thread: implementing rayleighchan in OFDM
- Next by thread: Re: symmetric component-wise perturbation
- Index(es):
Relevant Pages
|