• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.21 no.7
• /
• pp.805-814
• /
• 2010

The Crank-Nicolson isotropic-dispersion finite difference time domain(CN ID-FDTD) scheme is proposed based on isotropic-dispersion finite difference(ID-FD) $equation^{[1],[2]}$. The dispersion relation of CN ID-FDTD is derived for lossy media by solving the eigenvalue problem of iteration matrix in spatial spectral domain, in addition, the weighting factors and scaling factors of the CN ID-FDTD scheme are presented for low dispersion error. The CN ID-FDTD scheme makes the dispersion error drastically reduced and shows accurate numerical results compared to the conventional Crank-Nicolson FDTD method.

• Proceedings of the Korea Electromagnetic Engineering Society Conference
• /
• 2005.11a
• /
• pp.183-186
• /
• 2005

A two dimensional (2-D) finite-difference time-domain (FDTD) method based on a novel finite difference scheme is developed to eliminate the numerical dispersion errors. In this paper, numerical dispersion and stability analysis of the new scheme are given, which show that the proposed method is nearly dispersionless, and stable for a larger time step than the standard FDTD method.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.18 no.6 s.121
• /
• pp.664-673
• /
• 2007

In this paper, the ID-FDTD scheme, proposed in Reference [1] and [2], is modified and completely analyzed. The modifications are composed of three parts: rigorous stability analysis, dispersion relation for linear lossy media, and new scaling factors for permittivity, permeability, and conductivity. As a result, it is shown that the proposed scheme has lower dispersion error in spite of larger time step than the conventional standard scheme of Reference [3]. To validate the scheme, there are presented two scattering examples, which show excellent results.

• Journal of electromagnetic engineering and science
• /
• v.8 no.4
• /
• pp.139-143
• /
• 2008

In this paper, a new scheme to calculate the weighting factor of the 2-D isotropic-dispersion finite difference time domain(ID-FDTD) is proposed. The weighting factor in [1] was formulated in free space, so that it may not be optimal in dielectric media. Therefore, the weighting factor was reformulated by considering the material properties and using the least mean square method. As a result, a minimum numerical dispersion error for any dielectric media is guaranteed.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.17 no.2 s.105
• /
• pp.165-170
• /
• 2006

A two dimensional(2-D) finite-difference time-domain(FDTD) method based on a novel finite difference scheme is developed to eliminate the numerical dispersion errors. In this paper, numerical dispersion and stability analysis of the new scheme are given, which show that the proposed method is nearly dispersionless, and stable for a larger time step than the standard FDTD method.

• Journal of electromagnetic engineering and science
• /
• v.13 no.3
• /
• pp.158-164
• /
• 2013

This paper expands a previously proposed optimized higher order (2, 4) finite-difference time-domain scheme (H(2, 4) scheme) for use with lossy material. A low dispersion error is obtained by introducing a weighting factor and two scaling factors. The weighting factor creates isotropic dispersion, and the two scaling factors dramatically reduce the numerical dispersion error at an operating frequency. In addition, the results confirm that the proposed scheme performs better than the H(2, 4) scheme for wideband analysis. Lastly, the validity of the proposed scheme is verified by calculating a scattering problem of a lossy circular dielectric cylinder.

• Journal of electromagnetic engineering and science
• /
• v.10 no.2
• /
• pp.45-49
• /
• 2010

This paper presents an analytical formulation of the numerical reflection coefficient of the ID-FDTD scheme at the planar dielectric boundary for a TM wave incidence. The reflection coefficient is formulated in an approximate manner, and the accuracy of this method is numerically verified. The effective dielectric constant for a grid on the interface is obtained, and then reduced to that of the Yee scheme for a small cell size.