GPU Accelerated DGTD Method for Analyzing Electromagnetic Scattering Problems

In this paper, we propose an efficient method which use the GPU parallel technology to speed up the time iterative process of the time-domain discontinuous Galerkin method (DGTD). the parallelization strategy after introducing the Compute Unified Device Architecture (CUDA) devices is the focus of this paper. The numerical results prove that the DGTD not only has reliable accuracy but also has natural parallel advantages in the field of parallel technology.