Abstract
An efficient parallel computation using graphics processing units (GPUs) is developed for studying the electromagnetic (EM) backscattering characteristics from a large three-dimensional sea surface. A slope-deterministic composite scattering model (SDCSM), which combines the quasi-specular scattering of Kirchhoff Approximation (KA) and Bragg scattering of the two-scale model (TSM), is utilized to calculate the normalized radar cross section (NRCS in dB) of the sea surface. However, with the improvement of the radar resolution, there will be millions of triangular facets on the large sea surface which make the computation of NRCS time-consuming and inefficient. In this paper, the feasibility of using NVIDIA Tesla K80 GPU with four compute unified device architecture (CUDA) optimization strategies to improve the calculation efficiency of EM backscattering from a large sea surface is verified. The whole GPU-accelerated SDCSM calculation takes full advantage of coalesced memory access, constant memory, fast math compiler options, and asynchronous data transfer. The impact of block size and the number of registers per thread is analyzed to further improve the computation speed. A significant speedup of 748.26x can be obtained utilizing a single GPU for the GPU-based SDCSM implemented compared with the CPU-based counterpart performing on the Intel(R) Core(TM) i5-3450.
Highlights
Studying the electromagnetic (EM) backscattering characteristics from an electrically large sea surface plays an important role in the synthetic aperture radar (SAR) imaging, ocean parameter inversion, targets detection, and monitoring [1,2,3]
Based on the two-scale oceanic surface model simulated by the double superimposition model (DSM) [4,5], the Kirchhoff approximation (KA) [6,7], small perturbation method (SPM) [8], small-slope approximation method (SSA) [9,10], two-scale method [11], four-modified two-scale method (FMTSM) [12], and slope-deterministic composite scattering model (SDCSM) [13] can be quoted to calculate the normalized radar cross section (NRCS) from sea surface
Tesla K80 graphics processing units (GPUs) with five compute unified device architecture (CUDA) optimization strategies has been utilized to speed up the generation of the time-evolving oceanic surface model (TOSM) and a significant 791.0× speedup has been achieved compared to a serial C program
Summary
Studying the electromagnetic (EM) backscattering characteristics from an electrically large sea surface plays an important role in the synthetic aperture radar (SAR) imaging, ocean parameter inversion, targets detection, and monitoring [1,2,3]. Tesla K80 GPU with five CUDA optimization strategies has been utilized to speed up the generation of the time-evolving oceanic surface model (TOSM) and a significant 791.0× speedup has been achieved compared to a serial C program. A new GPU-based SDCSM implemented with five CUDA optimization strategies is developed for calculating the NRCS from an electrically large sea surface. This paper is organized as follows: first, based on the electrically large oceanic surface simulated by double superimposition model (DSM) [4,5], the SDCSM is exploited to calculate the NRCS and the results are compared with the experimental data to verify the correctness of the outcomes. After using diverse CUDA strategies, a significant speedup can be achieved by the GPU-based SDCSM algorithm executed on Tesla K80 compared with a serial C program executed on Intel(R) Core (TM) i5-3450 CPU
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
