Efficient backprojection-based synthetic aperture radar computation with many-core processors

Tackling computationally challenging problems with high efficiency often requires the combination of algorithmic innovation, advanced architecture, and thorough exploitation of parallelism. We demonstrate this synergy through synthetic aperture radar (SAR) via backprojection, an image reconstruction method that can require hundreds of TFLOPS. Computation cost is significantly reduced by our new algorithm of approximate strength reduction; data movement cost is economized by software locality optimizations facilitated by advanced architecture support; parallelism is fully harnessed in various patterns and granularities. We deliver over 35 billion backprojections per second throughput per compute node on an Intel® Xeon® processor E5-2670-based cluster, equipped with Intel® Xeon Phi coprocessors. This corresponds to processing a 3K×3K image within a second using a single node. Our study can be extended to other settings: backprojection is applicable elsewhere including medical imaging, approximate strength reduction is a general code transformation technique, and many-core processors are emerging as a solution to energy-efficient computing.

[1]  T. M. Benson,et al.  Gigapixel spotlight synthetic aperture radar backprojection using clusters of GPUs and CUDA , 2012, 2012 IEEE Radar Conference.

[2]  Stephen J. Wright,et al.  Hogwild: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent , 2011, NIPS.

[3]  Ahmed Fasih,et al.  GPU-accelerated synthetic aperture radar backprojection in CUDA , 2010, 2010 IEEE Radar Conference.

[4]  Lam H. Nguyen,et al.  Power versus performance tradeoffs of GPU-accelerated backprojection-based synthetic aperture radar image formation , 2011, Defense + Commercial Sensing.

[5]  G. Amdhal,et al.  Validity of the single processor approach to achieving large scale computing capabilities , 1967, AFIPS '67 (Spring).

[6]  Mehrdad Soumekh,et al.  Synthetic Aperture Radar Signal Processing with MATLAB Algorithms , 1999 .

[7]  Miriam Leeser,et al.  Parallel Backprojection: A Case Study in High-Performance Reconfigurable Computing , 2009, EURASIP J. Embed. Syst..

[8]  Eberhard Hansis,et al.  Iterative reconstruction for circular cone-beam CT with an offset flat-panel detector , 2010, IEEE Nuclear Science Symposuim & Medical Imaging Conference.

[9]  Eric J. Kelmelis,et al.  CULA: hybrid GPU accelerated linear algebra routines , 2010, Defense + Commercial Sensing.

[10]  D. Munson,et al.  A tomographic formulation of spotlight-mode synthetic aperture radar , 1983, Proceedings of the IEEE.

[11]  Ken Kennedy,et al.  An algorithm for reduction of operator strength , 1977, Commun. ACM.

[12]  Pradeep Dubey,et al.  3.5-D Blocking Optimization for Stencil Computations on Modern CPUs and GPUs , 2010, 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis.

[13]  W. Kenneth Jenkins,et al.  Convolution backprojection image reconstruction for spotlight mode synthetic aperture radar , 1992, IEEE Trans. Image Process..

[14]  M Grass,et al.  Motion-compensated iterative cone-beam CT image reconstruction with adapted blobs as basis functions , 2008, Physics in medicine and biology.

[15]  Michael Wolfe,et al.  High performance compilers for parallel computing , 1995 .

[16]  Charles V. Jakowatz,et al.  Beamforming as a foundation for spotlight-mode SAR image formation by backprojection , 2008, SPIE Defense + Commercial Sensing.

[17]  Ümit V. Çatalyürek,et al.  Investigating the use of GPU-accelerated nodes for SAR image formation , 2009, 2009 IEEE International Conference on Cluster Computing and Workshops.

[18]  Jean-Michel Muller,et al.  Elementary Functions: Algorithms and Implementation , 1997 .

[19]  Charles V. Jakowatz,et al.  An implementation of a fast backprojection image formation algorithm for spotlight-mode SAR , 2008, SPIE Defense + Commercial Sensing.

[20]  Jack E. Volder.  The CORDIC Trigonometric Computing Technique , 1959, IRE Trans. Electron. Comput..

[21]  Shu Xiao,et al.  An N2logN back-projection algorithm for SAR image formation , 2000, Conference Record of the Thirty-Fourth Asilomar Conference on Signals, Systems and Computers (Cat. No.00CH37154).

[22]  Charles V. Jakowatz,et al.  Spotlight-Mode Synthetic Aperture Radar: A Signal Processing Approach , 1996 .

[23]  Mark A. Richards,et al.  Fundamentals of Radar Signal Processing , 2005 .

[24]  Pradeep Dubey,et al.  Designing and dynamically load balancing hybrid LU for multi/many-core , 2011, Computer Science - Research and Development.

[25]  Alfred V. Aho,et al.  Compilers: Principles, Techniques, and Tools , 1986, Addison-Wesley series in computer science / World student series edition.