Electromagnetic Scattering Using Gpu-Based Finite Difference Frequency Domain Method

Zainud-Deen, S.H.; Hassan, Emad S.; Ibrahim, Mourad S.; Awadalla, K. H.; Botros, A.Z.

doi:10.2528/pierb09060703

Cited by 28 publications

(19 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GPU computing is being exploited in many scientific applications [6][7][8] with interesting results in the EM field and nanotechnology [9,10]. Much published work on EM computational simulations using the GPUs confirms the growing interest in this powerful technology of the computational EM community [11][12][13].…”

Section: Introductionmentioning

confidence: 86%

Gpu Approach for Hertzian Potential Formulation Tool Oriented on Electromagnetic Nanodevices

Tartarini¹,

Massaro²

2011

PIER M

View full text Add to dashboard Cite

Abstract-The time domain modeling and simulation of electromagnetic (EM) waves interaction with nanodevices, at high spatial and time resolution, requires high computational power. For the first time, in this paper we present an effective implementation of the Hertzian Potential Formulation (HPF) on the Graphics Processing Units (GPUs), through the NVIDIA's CUDA (Compute Unified Device Architecture) programming model. It accelerates the nanodevice EM simulations at nanometer scale harnessing the massive parallelism of the GPU based systems. This study is useful for similar electromagnetic codes including the Finite Difference approaches. The results demonstrate that this GPU tool outperforms the CPU based HPF implementation, reaching a speedup from 30 × to 70 ×.

show abstract

Section: Introductionmentioning

confidence: 86%

Gpu Approach for Hertzian Potential Formulation Tool Oriented on Electromagnetic Nanodevices

Tartarini¹,

Massaro²

2011

PIER M

View full text Add to dashboard Cite

show abstract

“…On the other hand, the most popularly used, FDTD method and its variants are not suited for modeling complex waveguide devices with narrow-band signal. The less popular FD-FD method and its variants [37][38][39][40][41], due to their need for solving large sparse linear equations, are more accurate in frequency domain applications.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Perpendicular Crossing Dielectric Waveguides With Various Typical Index Contrasts and Intersection Profiles

Chang¹,

Wu²

2010

PIER

View full text Add to dashboard Cite

Abstract-We present a rigorous 2D numerical study of the transmission, reflection and crosstalk coefficients of the perpendicular, identical dielectric crossing waveguide with various core-cladding index contrasts for both TE and TM polarizations. Our method is based on a hybrid frequency-domain finite-difference (FD-FD) technique computed with the cross-symmetry model. By varying the intersection profile, such as the circular, filleted, tapered and elliptical shapes, we achieve, even for a large 3.5 to 1.5 index ratio, a low 0.25 dB insertion loss, a nontrivial reduction over the straight direct crossing case.

show abstract

“…In particular, the GPU acceleration has recently become a hot topic of investigations due to its low cost in comparison to the highperformance computing on clusters. Several numerical methods have already been implemented on GPUs, i.e., FDTD method [18][19][20][21][22][23], finite-difference frequency-domain method [24,25], finite element method [26][27][28][29] and, method of moments [30][31][32].…”

Section: Introductionmentioning

confidence: 99%

Implementation of FDTD-Compatible Green's Function on Heterogeneous Cpu-Gpu Parallel Processing System

Stefański¹

2013

PIER

View full text Add to dashboard Cite

Abstract-This paper presents an implementation of the FDTDcompatible Green's function on a heterogeneous parallel processing system.The developed implementation simultaneously utilizes computational power of the central processing unit (CPU) and the graphics processing unit (GPU) to the computational tasks best suited for each architecture. Recently, closed-form expression for this discrete Green's function (DGF) was derived, which facilitates its applications in the FDTD simulations of radiation and scattering problems. Unfortunately, implementation of the new DGF formula in software requires a multiple precision arithmetic and may cause long runtimes. Therefore, an acceleration of the DGF computations on a CPU-GPU heterogeneous parallel processing system was developed using the multiple precision arithmetic and the OpenMP and CUDA parallel programming interfaces. The method avoids drawbacks of the CPUand GPU-only accelerated implementations of the DGF, i.e., long runtime on the CPU and significant overhead of the GPU initialization respectively for long and short length of the DGF waveform. As a result, the sevenfold speedup was obtained relative to the reference DGF implementation on a multicore CPU thus applicability of the DGF in FDTD simulations was significantly improved.

show abstract

Electromagnetic Scattering Using Gpu-Based Finite Difference Frequency Domain Method

Cited by 28 publications

References 14 publications

Gpu Approach for Hertzian Potential Formulation Tool Oriented on Electromagnetic Nanodevices

Gpu Approach for Hertzian Potential Formulation Tool Oriented on Electromagnetic Nanodevices

Analysis of Perpendicular Crossing Dielectric Waveguides With Various Typical Index Contrasts and Intersection Profiles

Implementation of FDTD-Compatible Green's Function on Heterogeneous Cpu-Gpu Parallel Processing System

Contact Info

Product

Resources

About