Gigapixel spotlight synthetic aperture radar backprojection using clusters of GPUs and CUDA

Abstract

Synthetic aperture radar (SAR) image formation via backprojection offers a robust mechanism by which to form images on general, non-planar surfaces, without often restrictive assumptions regarding the planarity of the wavefront at the locations being imaged. However, backprojection presents a substantially increased computational load relative to other image formation algorithms that typically depend upon fast Fourier transforms. In this paper, we present an image formation framework for accelerated SAR backprojection that utilizes a cluster of computing nodes, each with one or more graphics processing units (GPUs). We address the parallelization of the backprojection process among multiple nodes and the scalability thereby obtained, several optimization approaches, and performance as a function of both allocated resources and desired precision. Finally, we demonstrate the achieved performance on a simulated gigapixel-scale data set.

Efficient Backprojection-Based Synthetic Aperture Radar Computation with Many-Core Processors

“…where a x , a y , b x , b y and c xy are constants involving Δ x , Δ y , and first and second partial derivatives of f . 4 for m from 0 to M -1 for from 0 to L -1 4 We approximate around = 0 and m = 0 for illustrative purposes, instead of using a more accurate approximation around = L/2 and m = M/2. This more accurate approximation is a straightforward modification from the one presented here and is used in the evaluation section.z…”

“…Their partitioning method is aligned with ours -partition in the output image dimension first, and in the input pulse dimension only when it is necessary. There also have been GPU implementations of SAR backprojection [4,10,12,24]. GPUs provide hardware texture lookup support that improves the backprojection performance by Table 5 The projection of largest inputs that can be handled while satisfying the real-time constraint up to 15% at the expense of lower accuracy [4].…”

“…There also have been GPU implementations of SAR backprojection [4,10,12,24]. GPUs provide hardware texture lookup support that improves the backprojection performance by Table 5 The projection of largest inputs that can be handled while satisfying the real-time constraint up to 15% at the expense of lower accuracy [4]. GPUs also provide hardware trigonometric units but their performance benefits depend on required image formation accuracy, which will be explained later in this section.…”

GPU Processing for UAS-Based LFM-CW Stripmap SAR

“…Fortunately, backprojection processing is easily parallelized and computed efficiently on graphics processing units (GPU). Several studies have been conducted on implementing backprojection on GPUs, most notably Fasih and Hartley (2010), Benson et al, (2012), Capozzoli et al (2013), and Nguyen et al (2004), but these papers focus on the simplest form of spotlight mode SAR backprojection and are not directly applicable to stripmap SAR. In this paper we present the development and implementation of a highly efficient GPU-based SAR backprojection processor for stripmap imaging.…”

Acceleration of fast factorized back projection algorithm for bistatic SAR

“…Fortunately, now GPU processor has been used more and more widely in the field of scientific computing due to its powerful parallel processing capability. Some works about the using of GPUs for accelerating SAR direct backprojection have been reported [4]- [6] and indicate that it is beneficial significantly to shorten the time of SAR data processing.…”