I. Introduction
Since it has been introduced by Yee [1], the second-order accurate in both time and space finite difference time domain (FDTD) method has been used to numerically solve a wide range of electromagnetic problems [2]. The FDTDbased algorithms are advantageous through the simplicity of their computation and implementation. Besides, they are very suitable for modeling inhomogeneous geometries. However standard FDTD-based algorithms suffer from the numerical dispersion caused by the low order accuracy in both time and space domains. This fact provokes the necessity of using a fine grid to achieve satisfactory results and consequently a high cost regarding the utilization of the memory and the time of simulation [3].