perfectly matched layer

Try the UPML based approach, its much easier to comprehend and implement.

Thanks Hicham,

have you any links about UPML?

aze

for the UPML there is an entire chapter in the book 'advances in computational electrodynamics'

Also,look for papers by S.D.Gedney, he is the one that suggested this formulation and wrote the above chapter.

Papers by D.Sullivan might also help u to get the feel of such an approach. He proposed a similar unsplit method.Also his book on FDTD has a sample code (wih a few mistakes i have to warn u) but again it'll help u get an accurte picture of these approaches

Hi settmar,

I can't get the book nor the paper,
would you like please to upload them.

Thanks

Gedney's paper here

I succed to uilt a 3D FDTD code using UPML so if you are interested just contact me on this mal / microwave_tetuan@yahoo.com

Hi helio1972
At the moment I'm building a 2D FDTD code using UPML in MATLAB, but it is not working. I did check everything, but still cannot find where the problem is. So can you please upload your version for me to check with? Thank you very much!

Hi,
I have tried to write a matlab code based on TF/SF method 2D FDTD with Berenger's PML. I calculated the fields building a look-up table specified in Taflove's book.
I faced three problems when I run the simulation;
1) There are a lot of reflections (10% of the magnitude of incident field) in the computation space when there is no scatterer.

2) The TF region leaks the field in SF region.

3)When I place a PEC plane in the computation space, the H fields on the edge of the PEC plate never dissappears.

Have any idea what I am doing wrong?

Thanks

Hi guys,

Currently, I am working on the Berenger's PML. I am using Sullivan's book as my reference and I managed 2D program in Fortran. However, I have difficult to do it in 3D. Could someone gives any advise?

Anyone has the 3D code in either Matlab or Fortran? I can be contacted emkenkyu@yahoo.com

TQ