[scilab-Users] Usage of expm function

[Kirill Kirillov]

Hi, all.

I am using Tylor series method for matrix exponential computation. But
after reading [1] I'd like to switch to better method.

MATLAB uses a scaling and squaring algorithm with a Pade approximation,
which is described in [1] as well.

Scilab uses first block-diagonalizing and then applying a Pade
approximation on each block.

1. Should I use Scilab expm function or MATLAB algorithm?
2. Which file contains source code for Scilab expm function?
3. Where I can find description for Scilab algorithm.

Thanks.

[1] http://www.cs.cornell.edu/cv/researchpdf/19ways+.pdf

-- 
Kirill Kirillov
<><