Optimum time-step size for 2D (2, 4) FDTD method

The correct choice of time-step size is of crucial importance in the case of the (2, 4) finite-difference time-domain (FDTD) method, as it affects the algorithm´s overall accuracy and convergence rate. A description of the inherent discretisation error is introduced, which is exploited for the derivation of a simple, yet reliable, approximation of the optimum temporal sampling density. As the proposed approach attempts to remedy inaccuracies in a mean-value sense, it is shown that highly efficient calculations can be carried out in this way.