Highly efficient synthetic aperture radar processing system for airborne sensors using CPU+GPU architecture

Wu, Zheng; Liu, Yabo; Zhang, Lei; Li, Ning; Du, Kangning; Balz, Timo

doi:10.1117/1.jrs.9.097293

Cited by 5 publications

(6 citation statements)

References 19 publications

(14 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the help of CUDA technology, those traditional code implements large amounts of floating point computation with CPU can be easily ported to GPU to realize massive parallelism. The existing results of research indicate that GPU based method can improve the efficiency of SAR imaging by dozens of times [ 3 , 11 , 26 , 27 , 28 , 29 , 30 ], which outperforms the traditional HPC methods. However, there are still some further optimizations for the existing methods to accelerate SAR imaging processing.…”

Section: Introductionmentioning

confidence: 99%

“…For example, the issue of data transferring is not solved in Song’s method [ 11 ]. In addition, just 2 Tesla

cards are used in current papers [ 11 , 27 ]. Though the GPU based method can process huge amount of raw data and boost the computing efficiency many times, they still cannot meet the practical requirement for fast or even real-time SAR imaging.…”

Section: Introductionmentioning

confidence: 99%

“…Typically, CPU only takes on tasks of logical processing, controlling and data input/output (IO) [ 3 , 11 ]. Then other researchers further improve the GPU method by exploiting more computing power of CPU, which is realized by the multi-core parallelism [ 27 ]. The efficiency of multi-core CPU and GPU are in the different level, which makes the contribution of CPU computing unimpressive.…”

Section: Introductionmentioning

confidence: 99%

“…Compared to the existing research, we mainly make the following contributions: A bi-direction partitioning strategy is proposed to solve the contradiction between big raw data and limited GPU memory. It not only provides support for the multiple devices based parallel imaging, but also avoids the complex sub-aperture processing [ 27 ], which is employed by existing multi-GPU based

imaging to solve the aforementioned contradiction; A vector processing based multi-core CPU parallel imaging method is presented. Through the fine-grained vectorization parallel and coarse-grained multi-core parallel, the computing power of CPU is strengthened, and the gap between CPU and GPU computing is greatly narrowed; A deep multiple CPU/GPU collaborative computing method is presented to accelerate SAR imaging processing.…”

Section: Introductionmentioning

confidence: 99%

“…A bi-direction partitioning strategy is proposed to solve the contradiction between big raw data and limited GPU memory. It not only provides support for the multiple devices based parallel imaging, but also avoids the complex sub-aperture processing [ 27 ], which is employed by existing multi-GPU based

imaging to solve the aforementioned contradiction;…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Accelerating Spaceborne SAR Imaging Using Multiple CPU/GPU Deep Collaborative Computing

Zhang

et al. 2016

Sensors

View full text Add to dashboard Cite

With the development of synthetic aperture radar (SAR) technologies in recent years, the huge amount of remote sensing data brings challenges for real-time imaging processing. Therefore, high performance computing (HPC) methods have been presented to accelerate SAR imaging, especially the GPU based methods. In the classical GPU based imaging algorithm, GPU is employed to accelerate image processing by massive parallel computing, and CPU is only used to perform the auxiliary work such as data input/output (IO). However, the computing capability of CPU is ignored and underestimated. In this work, a new deep collaborative SAR imaging method based on multiple CPU/GPU is proposed to achieve real-time SAR imaging. Through the proposed tasks partitioning and scheduling strategy, the whole image can be generated with deep collaborative multiple CPU/GPU computing. In the part of CPU parallel imaging, the advanced vector extension (AVX) method is firstly introduced into the multi-core CPU parallel method for higher efficiency. As for the GPU parallel imaging, not only the bottlenecks of memory limitation and frequent data transferring are broken, but also kinds of optimized strategies are applied, such as streaming, parallel pipeline and so on. Experimental results demonstrate that the deep CPU/GPU collaborative imaging method enhances the efficiency of SAR imaging on single-core CPU by 270 times and realizes the real-time imaging in that the imaging rate outperforms the raw data generation rate.

show abstract

Section: Introductionmentioning

confidence: 99%

“…For example, the issue of data transferring is not solved in Song’s method [ 11 ]. In addition, just 2 Tesla

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

imaging to solve the aforementioned contradiction;…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Accelerating Spaceborne SAR Imaging Using Multiple CPU/GPU Deep Collaborative Computing

Zhang

et al. 2016

Sensors

View full text Add to dashboard Cite

show abstract

Scene optimization of GPU-based back-projection algorithm

et al. 2022

View full text Add to dashboard Cite

On-chip data organization and access strategy for spaceborne SAR real-time imaging processor

et al. 2021

View full text Add to dashboard Cite

Spaceborne SAR(synthetic aperture radar) imaging requires real-time processing of enormous amount of input data with limited power consumption. Designing advanced heterogeneous array processors is an effective way to meet the requirements of power constraints and real-time processing of application systems. To design an efficient SAR imaging processor, the on-chip data organization structure and access strategy are of critical importance. Taking the typical SAR imaging algorithm-chirp scaling algorithm-as the targeted algorithm, this paper analyzes the characteristics of each calculation stage engaged in the SAR imaging process, and extracts the data flow model of SAR imaging, and proposes a storage strategy of cross-region cross-placement and data sorting synchronization execution to ensure FFT/IFFT calculation pipelining parallel operation. The memory wall problem can be alleviated through on-chip multi-level data buffer structure, ensuring the sufficient data providing of the imaging calculation pipeline. Based on this memory organization and access strategy, the SAR imaging pipeline process that effectively supports FFT/IFFT and phase compensation operations is therefore optimized. The processor based on this storage strategy can realize the throughput of up to 115.2 GOPS, and the energy efficiency of up to 254 GOPS/W can be achieved by implementing 65 nm technology. Compared with conventional CPU+GPU acceleration solutions, the performance to power consumption ratio is increased by 63.4 times. The proposed architecture can not only improve the real-time performance, but also reduces the design complexity of the SAR imaging system, which facilitates excellent performance in tailoring and scalability, satisfying the practical needs of different SAR imaging platforms.

show abstract

Highly efficient synthetic aperture radar processing system for airborne sensors using CPU+GPU architecture

Cited by 5 publications

References 19 publications

Accelerating Spaceborne SAR Imaging Using Multiple CPU/GPU Deep Collaborative Computing

Accelerating Spaceborne SAR Imaging Using Multiple CPU/GPU Deep Collaborative Computing

Scene optimization of GPU-based back-projection algorithm

On-chip data organization and access strategy for spaceborne SAR real-time imaging processor

Contact Info

Product

Resources

About