Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Time domain physical optics algorithm based on CPU (Central Processing Unit) and GPU (Graphics Processing Unit) hybrid asynchronous parallel way

A physical optics, time domain technology, applied in computing, electrical digital data processing, special data processing applications, etc., can solve problems such as single platform, increased programming and debugging difficulty, and large space occupation, saving time and reducing development and writing costs , the effect of increasing the acceleration ratio

Active Publication Date: 2016-01-13
XIDIAN UNIV
View PDF4 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these parallel TDPOs are all implemented based on the CPU, and the platform is single. In order to achieve a high acceleration ratio, a high-performance workstation or even a PC cluster is often required, which occupies a large space and is expensive, and the cost of the simulation hardware is high. Due to the large number of CPUs, programming and debugging Difficulty increases, and there is an urgent need for a simple and efficient parallel programming method and platform to reduce the cost of software development and hardware simulation

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Time domain physical optics algorithm based on CPU (Central Processing Unit) and GPU (Graphics Processing Unit) hybrid asynchronous parallel way
  • Time domain physical optics algorithm based on CPU (Central Processing Unit) and GPU (Graphics Processing Unit) hybrid asynchronous parallel way
  • Time domain physical optics algorithm based on CPU (Central Processing Unit) and GPU (Graphics Processing Unit) hybrid asynchronous parallel way

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0040] The present invention will be described in further detail below in conjunction with the accompanying drawings.

[0041] The invention belongs to the field of numerical analysis of electromagnetic fields, and relates to the Gaussian integral formula in mathematical numerical analysis, which is an efficient algorithm that is easy to understand and implement for solving the integral equation of the time-domain physical optics equation without original functions. Parallel programming) accelerates the reading of model files and incident field information in parallel, combined with OpenMP (CPU shared memory parallel) parallel mode to accelerate file preprocessing, combined with OpenACC (GPU application programming interface) parallel mode to accelerate numerical calculations, which can be used to accelerate a large number of In terms of the RCS (radar cross section) of the electric large target and the real-time RCS (radar cross section) estimation of the electric large target...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a time domain physical optics algorithm based on a CPU (Central Processing Unit) and GPU (Graphics Processing Unit) hybrid asynchronous parallel way. The time domain physical optics algorithm comprises the following steps: 1, performing modeling through 3Dmax, and performing subdivision through triangular surface elements, wherein a derived model is in an OBJ format; 2, synchronously reading usable information such as vertex coordinates and a vertex quantity of the triangular surface elements in a model file (1), vertex numbers and a surface element quantity of the surface elements in the model file (2) and an incident electric field (3) respectively on three threads in an MPI parallel way; 3, accelerating the whole process in parallel through OpenMP; 4, transmitting data such as a Gaussian node array and a triangular surface element array into a GPU, accelerating a Gaussian integral value operation by the GPU to obtain a scattered field, and transmitting the scattered field into a CPU; and 5, transforming a scattered field time domain into a frequency domain through a Fourier transform, and dividing the frequency domain by an incident electric field frequency domain to obtain an RCS (Radar Cross Section) array. Through adoption of the time domain physical optics algorithm, time is saved greatly when a transient scattering calculation amount of an electrically large object is large.

Description

technical field [0001] The invention belongs to the field of electromagnetic field numerical analysis, relates to a Gaussian integral formula in mathematical numerical analysis, and specifically relates to a time-domain physical optics algorithm based on a mixed asynchronous parallel mode of CPU and GPU. Background technique [0002] The wide application of ultra-wideband radar and microwave and millimeter-wave radar has made the research on the wide-band time-domain characteristics of electrically large-sized targets attract extensive attention. Physical optics approximation is one of the efficient methods to analyze the frequency-domain electromagnetic properties of electrically large objects. However, only limited physical optics analyzes of transient or pulsed excitations in the time domain have been reported. In the early time-domain physical optics approximation, the frequency-domain response of the target excited by the time-harmonic source was obtained first, and th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F19/00
Inventor 徐乐赵伟李蕊史小卫
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products