Video-SAR Implementation on Various Computing Architectures

Forming synthetic aperture radar images in real-time using the backprojection algorithm (BPA) requires optimal use of available hardware. We discuss the challenges and feasibility of implementing the BPA to achieve good performance on graphics processor units (GPU) platforms. We will compare the ach...

Full description

Saved in:

Bibliographic Details
Published in:	2024 International Radar Symposium (IRS) pp. 44 - 49
Main Authors:	Fritzenwallner, Michael, Stelzer, Andreas, Haderer, Andreas, Feger, Reinhard
Format:	Conference Proceeding
Language:	English
Published:	Warsaw University of Technology 02-07-2024
Subjects:	Backprojection FMCW GPU Graphics processing units Hardware Image edge detection Performance evaluation Radar tracking Real-time systems Runtime SAR
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Forming synthetic aperture radar images in real-time using the backprojection algorithm (BPA) requires optimal use of available hardware. We discuss the challenges and feasibility of implementing the BPA to achieve good performance on graphics processor units (GPU) platforms. We will compare the achieved performances between high-end GPUs, CPUs (x86-64 and ARM) as well as edge devices with low-performance GPUs. Benchmark results prove the feasibility of real-time processing with embedded GPUs using less than 40W of power.
ISSN:	2155-5753