Three-dimensional angle-optimized FDTD algorithms for large-scale problems

We describe a 3D angle-optimized FDTD (AO-FDTD) algorithm to minimize the dispersion error around pre-assigned angles. Because of the 3D Cartesian grid symmetry, the optimization occurs at each octant concurrently. This algorithm is of particular interest for some large-scale FDTD simulations where the dispersion error around a limited angular span should be minimized, while the overall isotropy is not critical.