Parallel Implementation of a 3d Subgridding FDTD Algorithm for Large Simulations

Vaccari, Alessandro; Lesina, Antonino Calà; Cristoforetti, Luca; Pontalti, R.

doi:10.2528/pier11063004

Cited by 38 publications

(38 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is worth to note that the highresolution standard FDTD of the whole area (entitled FDTD), which is used as the reference, modeled the completely computational domain. To overcome the memory limit of a serial processor, the parallel implementation is used [27][28][29][30]. The convolution PML is used to truncate the computational domain in this work [31][32][33].…”

Section: Numerical Resultsmentioning

confidence: 99%

A Simple Local Approximation FDTD Model of Short Apertures With a Finite Thickness

Xiong¹,

Chen²,

Mao³

et al. 2012

PIER

View full text Add to dashboard Cite

Abstract-This paper brings forward a simple local approximation finite-difference time-domain (FDTD) method for the analysis of short apertures with a finite thickness. By applying the equivalence principle together with a simple local approximation, the varying field distribution is accurately derived. The updating equations for the slot field can be derived by casting the field distributions into the contour paths containing the apertures. The method has been applied to two problems and the results are compared with the high-resolution standard FDTD simulation results and the measurement results. The accuracy of the proposed method is verified from the comparison of both the field distribution and the time-domain and frequency-domain slot coupling results. It is demonstrated that the local approximation is highly efficient and timesaving, and the present method is stable, numerically and computationally efficient.

show abstract

Section: Numerical Resultsmentioning

confidence: 99%

A Simple Local Approximation FDTD Model of Short Apertures With a Finite Thickness

Xiong¹,

Chen²,

Mao³

et al. 2012

PIER

View full text Add to dashboard Cite

show abstract

“…Previous parallel algorithms (e.g., parallel FDTD [10,11], parallel direct solver [12] and parallel MLFMM [13,14]) were mainly implemented on CPU clusters. Due to the high performance/cost ratio and the fast performance growth of GPUs, GPU clusters are becoming more and more popular.…”

Section: Introductionmentioning

confidence: 99%

Parallel Shooting and Bouncing Ray Method on Gpu Clusters for Analysis of Electromagnetic Scattering

Tao¹,

Lin²

2013

PIER

View full text Add to dashboard Cite

Abstract-This paper proposes an efficient parallel shooting and bouncing ray (SBR) method on the graphics processing unit (GPU) cluster for solving the electromagnetic scattering problems. At each incident direction, the parallel SBR method partitions the virtual aperture into sub-apertures, and distributes the computational process of each sub-aperture over GPU nodes. As ray tubes in the virtual aperture do not have the same computational time, the parallel efficiency highly depends on how to partition the virtual aperture. This paper addresses this issue by a dynamic partitioning scheme according to the computational time at the previous angle, which can achieve excellent load balance. Numerical examples are presented to demonstrate the accuracy, high parallel efficiency, good scalability and versatility of the proposed method.

show abstract

“…Once the estimated optimal weighting factor a is obtained, a revision procedure on the anisotropic numerical space should be performed: 1) choose the fastest phase velocity, which propagates along the grid main diagonals, to be the exact one, thus no phase velocities exceed light speed c after revision; 2) equate the wave number k in (9) to k exact and substitute c with c/sf , where sf is the scaling factor; 3) work out the value of sf and revise the electromagnetic attributes with ε = sf × ε and µ = sf × µ.…”

Section: Determination Of Optimal Weighting Factormentioning

confidence: 99%

“…The finite-difference-time-domain (FDTD) method and its enhanced methods [1][2][3][4][5][6][7][8] are widely used in modeling electromagnetic problems due to their simpleness in updating equations and easiness in numerical implementation [9][10][11][12][13][14]. Constrained by the Courant-Friedrichs-Lewy (CFL) condition, however, its maximum time step is limited by the minimum cell size, which seriously affects its computational efficiency when fine meshes are required in the object under analysis [15].…”

Section: Introductionmentioning

confidence: 99%

Reduction of Numerical Dispersion of Adi-FDTD Method With Quasi Isotropic Spatial Difference Scheme

Yi-long¹,

Su²,

Liu³

2013

PIER B

View full text Add to dashboard Cite

Abstract-In this paper, the difference scheme of the alternatingdirection-implicit finite-difference time-domain (ADI-FDTD) method is replaced by the quasi isotropic (QI) spatial difference scheme to improve its numerical dispersion characteristics. The unconditional stability advantage of QI-ADI-FDTD is analytically proven and numerically verified. The numerical dispersion of the novel method can be dramatically reduced by choosing proper weighting factor. An example is simulated to demonstrate the accuracy and efficiency of the proposed method.

show abstract

Parallel Implementation of a 3d Subgridding FDTD Algorithm for Large Simulations

Cited by 38 publications

References 26 publications

A Simple Local Approximation FDTD Model of Short Apertures With a Finite Thickness

A Simple Local Approximation FDTD Model of Short Apertures With a Finite Thickness

Parallel Shooting and Bouncing Ray Method on Gpu Clusters for Analysis of Electromagnetic Scattering

Reduction of Numerical Dispersion of Adi-FDTD Method With Quasi Isotropic Spatial Difference Scheme

Contact Info

Product

Resources

About