giovanni6620
Newbie level 1
![Newbie level 1](/styles/images/ranks/alevel1.gif)
Hello,
hope someone can help me with this problem.
I'm a COMSOL user, and I usually simulate multiport problem; for what I've seen in my experience with this software, in order to obtain the system's S matrix, a RF simulation with n ports needs n times the solution time of the same model singularly excited.
As far as I know about FEM theory, there's not need to run the solver for every excitation, but it could be performed in only one step (i.e. matrix inversion); furthermore this is confirmed by HFSS sw if used for the same model.
Does anybody can help me with this doubt? Is it really necessary to run as much simulations as much is the number of excitations?
Thanks
Giovanni
hope someone can help me with this problem.
I'm a COMSOL user, and I usually simulate multiport problem; for what I've seen in my experience with this software, in order to obtain the system's S matrix, a RF simulation with n ports needs n times the solution time of the same model singularly excited.
As far as I know about FEM theory, there's not need to run the solver for every excitation, but it could be performed in only one step (i.e. matrix inversion); furthermore this is confirmed by HFSS sw if used for the same model.
Does anybody can help me with this doubt? Is it really necessary to run as much simulations as much is the number of excitations?
Thanks
Giovanni