Accelerated Implementation of the S-MRTD Technique Using Graphics Processor Units

Baron, G.S.; Fiume, Eugene; Sarris, Costas D.

doi:10.1109/mwsym.2006.249948

Cited by 9 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…M. Inman et al implemented the FDTD procedure on ATI Radeon x800 GPU and achieved 14 times increase in speed when compared to a 2.4 GHz computer, [38], [39]. On the other hand, Baron et al demonstrated that GPU is more suitable to implement MRTD than FDTD because of the limited memory resource on GPU hardware [40]; a detail description of the implementation is given in [41].…”

Section: Discussion and Outlookmentioning

confidence: 99%

Exploit the Parallel Paradigm

2010

IEEE Microwave

View full text Add to dashboard Cite

Section: Discussion and Outlookmentioning

confidence: 99%

Exploit the Parallel Paradigm

2010

IEEE Microwave

View full text Add to dashboard Cite

“…However, graphics hardware-based scientific computing is currently limited by the fixed amount of memory available in graphics cards. This paper investigates the graphics card implementation of the scaling multiresolution timedomain technique of [2], extending previous work reported by the authors in [16]. Thus, the issue of the limited memory resources available in a graphics card is addressed by replacing the FDTD scheme with a much more memory efficient one.…”

Section: Introductionmentioning

confidence: 88%

“…Under consideration was the same square domain employed in our preceding evaluation of accuracy. To evaluate the burden of mesh size on performance, here the domain was set to progressively larger dimensions of 0.16 Â 0 16,. 0.32 Â 0.32, 0.64 Â 0.64, 1.28 Â 1.28, and 2.56 Â 2.56 m 2 , spatially discretised at a constant rate of Dx ¼ Dz ¼ 0.0025 m. The resulting meshes of 64 Â 64, 128 Â 128, 256 Â 256, 512 Â 512 and 1024 Â 1024 cells progressively represent four-fold increases in memory overhead.…”

mentioning

confidence: 99%

Graphics hardware accelerated multiresolution time-domain technique: development, evaluation and applications

Baron

Fiume

Sarris

2008

IET Microw. Antennas Propag.

Self Cite

View full text Add to dashboard Cite

Recently, the use of graphics processing units as a means of achieving the hardware acceleration of the finite-difference time-domain (FDTD) technique has attracted significant interest in the computational electromagnetics community. However, the large memory requirements of the FDTD, compounded by the limited memory resources available in graphics processing units, compromise the efficiency of this approach. Alternatively, the authors show how the implementation of the multiresolution time-domain technique in a graphics processing unit can optimally utilise the memory resources of the latter and achieve unprecedented acceleration rates, significantly higher than those achievable by the FDTD. A detailed description of the proposed implementation is provided, followed by rigorous numerical error and performance evaluation studies that conclusively verify the advantages of the graphics accelerated multiresolution time domain. Finally, the potential of this technique as a fast microwave wireless channel modelling tool is demonstrated.

show abstract

“…Expansion of the FDTD differential formulation in space with the Deslauriers-Dubuc biorthogonal interpolating functions leads to the Scaling Multi-Resolution Time Domain (S-MRTD) method [63]- [65]. Accuracy remains comparable to the FDTD but at lower spatial sampling, which results in up to 8 times acceleration [64].…”

Section: Other Full-wave Techniques and Hybridsmentioning

confidence: 99%

Viability of Numerical Full-Wave Techniques in Telecommunication Channel Modelling

Novak

2020

J. commun. softw. syst. (Online)

View full text Add to dashboard Cite

In telecommunication channel modelling the wavelength is small compared to the physical features of interest, therefore deterministic ray tracing techniques provide solutions that are more efficient, faster and still within time constraints than current numerical full-wave techniques. Solving fundamental Maxwell's equations is at the core of computational electrodynamics and best suited for modelling electrical field interactions with physical objects where characteristic dimensions of a computing domain is on the order of a few wavelengths in size. However, extreme communication speeds, wireless access points closer to the user and smaller pico and femto cells will require increased accuracy in predicting and planning wireless signals, testing the accuracy limits of the ray tracing methods. The increased computing capabilities and the demand for better characterization of communication channels that span smaller geographical areas make numerical full-wave techniques attractive alternative even for larger problems. The paper surveys ways of overcoming excessive time requirements of numerical full-wave techniques while providing acceptable channel modelling accuracy for the smallest radio cells and possibly wider. We identify several research paths that could lead to improved channel modelling, including numerical algorithm adaptations for large-scale problems, alternative finite-difference approaches, such as meshless methods, and dedicated parallel hardware, possibly as a realization of a dataflow machine.

show abstract

Accelerated Implementation of the S-MRTD Technique Using Graphics Processor Units

Cited by 9 publications

References 7 publications

Exploit the Parallel Paradigm

Exploit the Parallel Paradigm

Graphics hardware accelerated multiresolution time-domain technique: development, evaluation and applications

Viability of Numerical Full-Wave Techniques in Telecommunication Channel Modelling

Contact Info

Product

Resources

About