The discrete Green's function (DGF) formulation of the finite-difference time-domain method (FDTD) is accelerated on a graphics processing unit (GPU) by means of the Compute Unified Device Architecture (CUDA) technology. In the developed implementation of the DGF-FDTD method, a new analytic expression for dyadic DGF derived based on scalar DGF is employed in computations. The DGF-FDTD method on GPU returns solutions that are compatible with the FDTD grid enabling the perfect hybridization of FDTD with the use of time-domain integral equation methods. The correctness of the results of the DGF-FDTD simulations on GPU is verified with the use of the FDTD method executed on a multicore central processing unit (CPU). The developed implementation provides maximally a six-fold speedup relative to the code executed on multicore CPU.
Authors
Additional information
- DOI
- Digital Object Identifier link open in new tab 10.1109/mikon.2016.7492112
- Category
- Aktywność konferencyjna
- Type
- materiały konferencyjne indeksowane w Web of Science
- Language
- angielski
- Publication year
- 2016
