A Parallel Block Splitting FFT Method for Efficient Computation of Electromagnetic Scattering from Time-Varying Sea Surface.

Saved in:

Bibliographic Details
Title:	A Parallel Block Splitting FFT Method for Efficient Computation of Electromagnetic Scattering from Time-Varying Sea Surface.
Authors:	Liu, Zhiwei¹, Xia, Feng¹ xfvxh2599@163.com, Wang, Rui¹, Zhang, Xiaoyan¹
Source:	Progress in Electromagnetics Research C. 2026, Vol. 165, p276-283. 8p.
Subjects:	Fast Fourier transforms, Parallel programming, Scalability, Digital computer simulation, Approximation error, Message passing (Computer science), Electromagnetic wave scattering, Ocean surface topography
Abstract:	2-D Fast Fourier transform (FFT) is the most time-consuming step for the modeling of time-varying sea surface using highorder small slope approximation (SSA). In this paper, a parallel block splitting method is proposed to accelerate 2D FFT calculation. The whole 2-D FFT matrix is divided into m x n blocks, and the traditional 2-D FFT is applied to each block in parallel. Finally, the complete FFT result can be obtained by using the message passing interface (MPI) for data communication and superimposing phase factors. This method can effectively reduce the communication overhead by combining symmetric domain decomposition and is more suitable than traditional parallel libraries. Accordingly, both generations of the sea surface and computation of scattering using SSA can be accelerated. Numerical experiments demonstrate that the proposed method exhibits strong scalability. Under a four-node configuration, the parallel efficiency of sea surface generation reaches 61.2%, while the second-order SSA parallel efficiency achieves 80.7%. This effectively resolves low-efficiency issues in large-scale sea surface generation and serial SSA computations. [ABSTRACT FROM AUTHOR]
	Copyright of Progress in Electromagnetics Research C is the property of Electromagnetics Academy and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	2-D Fast Fourier transform (FFT) is the most time-consuming step for the modeling of time-varying sea surface using highorder small slope approximation (SSA). In this paper, a parallel block splitting method is proposed to accelerate 2D FFT calculation. The whole 2-D FFT matrix is divided into m x n blocks, and the traditional 2-D FFT is applied to each block in parallel. Finally, the complete FFT result can be obtained by using the message passing interface (MPI) for data communication and superimposing phase factors. This method can effectively reduce the communication overhead by combining symmetric domain decomposition and is more suitable than traditional parallel libraries. Accordingly, both generations of the sea surface and computation of scattering using SSA can be accelerated. Numerical experiments demonstrate that the proposed method exhibits strong scalability. Under a four-node configuration, the parallel efficiency of sea surface generation reaches 61.2%, while the second-order SSA parallel efficiency achieves 80.7%. This effectively resolves low-efficiency issues in large-scale sea surface generation and serial SSA computations. [ABSTRACT FROM AUTHOR]
ISSN:	19378718
DOI:	10.2528/PIERC25121502