An efficient algorithm in time domain - ADI/R-FDTD

A new time domain algorithm (named ADI/R-FDTD), which combines two modified finite-difference time-domain (FDTD) methods (ADI-FDTD and R-FDTD), is proposed in this paper. The algorithm is derived from Maxwell's curl and divergence equations. It inherits the advantages of the former two methods, i.e., not only eliminating the restraint of the courant-friedrich-levy (C-F-L) condition and resulting in an efficient saving of CPU time, but also reaching a required memory reduction (up to about 1/3 of ADI-FDTD) in the storage of the field components. This paper presents a detailed description of the formulation and then makes a comparison between the simulation results of ADI/R-FDTD and those obtained from FDTD, ADI-FDTD and R-FDTD. And good agreement is achieved.