Title :
Acoustic wvave propagation modeling on 3D CNN-UM architecture
Author :
Sonkoly, P. ; Kozma, P. ; Nagy, Zsolt ; Szolgay, P.
Author_Institution :
Dept. of Image Process. & Neurocomput., Pannonia Univ., Veszprem
Abstract :
Solution of partial differential equations (PDE) has long been one of the most important fields of mathematics. Several previous studies proved the effectiveness of the CNN-UM solution for partial differential equations in two dimensions. This paper works with the 3D acoustic wave equation which describes the pressure wave propagation in fluid medium. This three-dimensional PDE can be solved also with 3D CNN-UM and multilayer 2D CNN-UM architecture. Unfortunately the huge number of space-dependent equations and the low computational precision do not make it possible to utilize the huge computing power of the analogue CNN-UM chips so the Falcon emulated digital CNN-UM architecture is used to implement our solution
Keywords :
acoustic wave propagation; cellular neural nets; partial differential equations; physics computing; 3D CNN-UM architecture; 3D acoustic wave equation; Falcon emulated digital CNN-UM architecture; acoustic wave propagation modeling; cellular neural networks; field programmable gate arrays; fluid medium; multilayer 2D CNN-UM architecture; partial differential equations; pressure wave propagation; Acoustic propagation; Acoustic waves; Analog computers; Cellular neural networks; Computer architecture; Computer networks; Differential equations; Electronic mail; Partial differential equations; Very large scale integration; Acoustic propagation; Cellular neural networks; Field programmable gate arrays;
Conference_Titel :
Cellular Neural Networks and Their Applications, 2006. CNNA '06. 10th International Workshop on
Conference_Location :
Istanbul
Print_ISBN :
1-4244-0639-0
Electronic_ISBN :
1-4244-0640-4
DOI :
10.1109/CNNA.2006.341611