This code follows the procedure of the FDFD method outlined in Dr. Raymond Rumpf's lecture notes from the EE 5337 course at UTEP. This code was written as a final project for EEL 6935 Linear Algebra Course at USF.
As of the latest commit, the solver is working for a 24 GHz binary diffraction grating using a slight modification to how the incident wave vector is being calculated. It is currently off by a scale of 0.0012.
- Julia 1.0
- IterativeSolvers
- SparseArrays
- LinearAlgebra
- Makie (for plotting)
Simply run the script
$ julia 2DFDFD.jl
I intend to discover why the k_inc section of the code is incorrect. Once I do that, I will add support for 3D solving and importing step file geometry. Included is an STL rasterizer in Utilities.jl. Additionally, I will add iterative convergance, a better iterative solver using Incomplete LU preconditioning, and calculations for transmitted and reflected power.