Finite-difference time-___domain method: Difference between revisions

Revision as of 12:20, 7 October 2005

Revision as of 14:32, 18 October 2005

Line 49:

FDTD finds the E/H fields directly everywhere in the computational ___domain. If the field values at some distance (like 10 meters away) are desired, it is likely that this distance will force the computational ___domain to be excessively large. Far field extensions are available for FDTD, but require some amount of post processing.

Since the simulation calulates the E and H fields at all points within the computational ___domain, it is best if the computational ___domain is finite. In many cases this is achieved by craetingcreating artificial boundaries into the simulation. Care must be taken to minimize errors introduced by such boundaries. There are a number of boundary conditions to chose from to simulate the effect of surrounding the computational ___domain with infinite free space.

Because FDTD is solved by propagating the fields forward

in the time ___domain, the time response of the medium through which they travel needs to be modelled explicitly. For arbitrary response, this will involve a computationally expensive convolution, although in most cases the time response (or dispersion) can be modelled more simply. An alternative way of solving [[Maxwell's_equations]] that can treat arbitrary dispersion easily is the Pseudospectral Spatial-Domain method ([[PSSD]]), which instead propagates the

fields forward in space.

== References ==