We present an ω-k approach based on the use of a 1D Non-Uniform FFT (NUFFT) routine, of NER (Non-Equispaced Results) type, programmed on a GPU by means of a Matlab code speeded-up by Accelereyes Jacket. The implementation of the GPU-based, parallel NUFFT routine is detailed and the computational advantages of the developed approach are highlighted against a multi-core CPU procedure implemented in Matlab language. The GPU-based routine has resulted about 5 times faster than the multicore CPU-based one. Experimental results of vegetation imaging on a box-tree (Buxus tree) are also presented.
Fast imaging of vegetation on GPUs based on Non-Uniform FFTs / Capozzoli, Amedeo; Curcio, Claudio; Iudice, I.; Liseno, Angelo; Savarese, S.. - (2011), pp. 1205-1208. (Intervento presentato al convegno International Conference on Electromagnetics in Advanced Applications tenutosi a Turin, Italy nel Sept. 12-16, 2011) [10.1109/ICEAA.2011.6046522].
Fast imaging of vegetation on GPUs based on Non-Uniform FFTs
CAPOZZOLI, AMEDEO;CURCIO, CLAUDIO;I. Iudice;LISENO, ANGELO;
2011
Abstract
We present an ω-k approach based on the use of a 1D Non-Uniform FFT (NUFFT) routine, of NER (Non-Equispaced Results) type, programmed on a GPU by means of a Matlab code speeded-up by Accelereyes Jacket. The implementation of the GPU-based, parallel NUFFT routine is detailed and the computational advantages of the developed approach are highlighted against a multi-core CPU procedure implemented in Matlab language. The GPU-based routine has resulted about 5 times faster than the multicore CPU-based one. Experimental results of vegetation imaging on a box-tree (Buxus tree) are also presented.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
