Convolution Models with Shift-invariant kernel based on Matlab-GPU platform for Fast Acoustic Imaging - CentraleSupélec Access content directly
Conference Papers Year : 2014

Convolution Models with Shift-invariant kernel based on Matlab-GPU platform for Fast Acoustic Imaging

Abstract

Acoustic imaging is an advanced technique for acoustic source localization and power reconstruc-tion from limited noisy measurements at microphone sensors. This technique not only involves in a forward model of acoustic propagation from sources to sensors, but also its numerical solution of an ill-posed inverse problem. Nowadays, the Bayesian inference methods in inverse methods have been widely investigated for robust acoustic imaging, but most of Bayesian methods are time-consuming, and one of the reasons is that the forward model causes heavy matrix multiplication. In this paper, we focus on the acceleration of the forward model by using a 2D-invariant convo-lution and a separable convolution respectively; For hardware acceleration, the Matlab-Graphics Processing Unit application are discussed. For method validation, we use the simulated and real data from the wind tunnel experiment in automobile industry.
Fichier principal
Vignette du fichier
ISAV2014.pdf (5.06 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01103819 , version 1 (15-01-2015)

Identifiers

  • HAL Id : hal-01103819 , version 1

Cite

Ning Chu, Nicolas Gac, José Picheral, Ali Mohammad-Djafari. Convolution Models with Shift-invariant kernel based on Matlab-GPU platform for Fast Acoustic Imaging. ISAV 2014, Dec 2014, Tehran, Iran. ⟨hal-01103819⟩
198 View
215 Download

Share

Gmail Facebook X LinkedIn More