嵌入式GPU滑动聚束SAR实时成像方法

Shan Qing Hu; Hui Xing Li; Bing Yi Li; Yi Zhuang Xie; Liang Chen; He Chen

doi:10.15918/j.tbit1001-0645.2019.056

嵌入式GPU滑动聚束SAR实时成像方法

Translated title of the contribution: The Real-Time Imaging Method for Sliding Spotlight SAR Based on Embedded GPU

Shan Qing Hu, Hui Xing Li, Bing Yi Li, Yi Zhuang Xie^*, Liang Chen, He Chen

^*Corresponding author for this work

School of Information and Electronics

Research output: Contribution to journal › Article › peer-review

11 Citations (Scopus)

Abstract

For the real-time SAR imaging systems, aiming at the problems of low real-time performance and high power consumption of traditional computing platforms, an implementation method was studied for embedded GPU. In order to make full use of the limited memory in the embedded GPU, a memory partitioning and reconfiguration scheme was proposed. The page-locked memory and zero-copy technology were used to realize the transmission-calculation parallelization. To achieve high real-time performance, large-scale parallelism was realized by using shared memory, registers, et. The result shows that sliding spotlight SAR imaging processing on the TX2 can only take 12.66 s time and consume 15 W power. This method is also applicable to other modes radar processing algorithms, and can provide reference for the future embedded real-time SAR imaging processing.

Translated title of the contribution	The Real-Time Imaging Method for Sliding Spotlight SAR Based on Embedded GPU
Original language	Chinese (Traditional)
Pages (from-to)	1018-1025
Number of pages	8
Journal	Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology
Volume	40
Issue number	9
DOIs	https://doi.org/10.15918/j.tbit1001-0645.2019.056
Publication status	Published - 1 Sept 2020

Access to Document

10.15918/j.tbit1001-0645.2019.056

Cite this

@article{d73564252e1a41569cb4921398304608,

title = "嵌入式GPU滑动聚束SAR实时成像方法",

abstract = "For the real-time SAR imaging systems, aiming at the problems of low real-time performance and high power consumption of traditional computing platforms, an implementation method was studied for embedded GPU. In order to make full use of the limited memory in the embedded GPU, a memory partitioning and reconfiguration scheme was proposed. The page-locked memory and zero-copy technology were used to realize the transmission-calculation parallelization. To achieve high real-time performance, large-scale parallelism was realized by using shared memory, registers, et. The result shows that sliding spotlight SAR imaging processing on the TX2 can only take 12.66 s time and consume 15 W power. This method is also applicable to other modes radar processing algorithms, and can provide reference for the future embedded real-time SAR imaging processing.",

keywords = "Embedded GPU, On-orbit real-time processing, Sliding spotlight, Synthetic aperture radar (SAR)",

author = "Hu, {Shan Qing} and Li, {Hui Xing} and Li, {Bing Yi} and Xie, {Yi Zhuang} and Liang Chen and He Chen",

year = "2020",

month = sep,

day = "1",

doi = "10.15918/j.tbit1001-0645.2019.056",

language = "繁体中文",

volume = "40",

pages = "1018--1025",

journal = "Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology",

issn = "1001-0645",

publisher = "Beijing Institute of Technology",

number = "9",

}

TY - JOUR

T1 - 嵌入式GPU滑动聚束SAR实时成像方法

AU - Hu, Shan Qing

AU - Li, Hui Xing

AU - Li, Bing Yi

AU - Xie, Yi Zhuang

AU - Chen, Liang

AU - Chen, He

PY - 2020/9/1

Y1 - 2020/9/1

N2 - For the real-time SAR imaging systems, aiming at the problems of low real-time performance and high power consumption of traditional computing platforms, an implementation method was studied for embedded GPU. In order to make full use of the limited memory in the embedded GPU, a memory partitioning and reconfiguration scheme was proposed. The page-locked memory and zero-copy technology were used to realize the transmission-calculation parallelization. To achieve high real-time performance, large-scale parallelism was realized by using shared memory, registers, et. The result shows that sliding spotlight SAR imaging processing on the TX2 can only take 12.66 s time and consume 15 W power. This method is also applicable to other modes radar processing algorithms, and can provide reference for the future embedded real-time SAR imaging processing.

AB - For the real-time SAR imaging systems, aiming at the problems of low real-time performance and high power consumption of traditional computing platforms, an implementation method was studied for embedded GPU. In order to make full use of the limited memory in the embedded GPU, a memory partitioning and reconfiguration scheme was proposed. The page-locked memory and zero-copy technology were used to realize the transmission-calculation parallelization. To achieve high real-time performance, large-scale parallelism was realized by using shared memory, registers, et. The result shows that sliding spotlight SAR imaging processing on the TX2 can only take 12.66 s time and consume 15 W power. This method is also applicable to other modes radar processing algorithms, and can provide reference for the future embedded real-time SAR imaging processing.

KW - Embedded GPU

KW - On-orbit real-time processing

KW - Sliding spotlight

KW - Synthetic aperture radar (SAR)

UR - http://www.scopus.com/inward/record.url?scp=85092687526&partnerID=8YFLogxK

U2 - 10.15918/j.tbit1001-0645.2019.056

DO - 10.15918/j.tbit1001-0645.2019.056

M3 - 文章

AN - SCOPUS:85092687526

SN - 1001-0645

VL - 40

SP - 1018

EP - 1025

JO - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

JF - Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

IS - 9

ER -

嵌入式GPU滑动聚束SAR实时成像方法

Abstract

Access to Document

Other files and links

Fingerprint

Cite this