DocumentCode :
507736
Title :
A Parallel Independent Component Implement Based on Field-Programmable-Gate-Array
Author :
Wang, Jing Hui ; Zhao, Yuan Chao
Author_Institution :
Tianjin Key Lab. of Intell. Comput. & Novel Software Technol., Tianjin Univ. of Technol., Tianjin, China
Volume :
2
fYear :
2009
fDate :
14-16 Aug. 2009
Firstpage :
500
Lastpage :
503
Abstract :
The independent component analysis (ICA) algorithm has been extensively investigated in its theory, implementation and applications. Up to now, there have many algorithms about ICA. The ICA algorithms can be roughly categorized as gradient-based, the second order gradient-based and diagonalization-based, etc. ICA algorithm requires a huge number of calculations. Our work wants to solve these limitations by increasing the computational speed. On a way of increasing the computational speed is using multiple IP cores work together for a single problem. In the paper, we designed parallel ICA IP core and completed the parallel ICA algorithm modules on Altera Stratix II GX devices.
Keywords :
blind source separation; field programmable gate arrays; gradient methods; independent component analysis; parallel algorithms; Altera Stratix II GX devices; diagonalization algorithm; field-programmable-gate-array; independent component analysis algorithm; parallel ICA IP core; parallel ICA algorithm module; parallel independent component; second order gradient algorithm; Algorithm design and analysis; Application software; Blind source separation; Concurrent computing; Covariance matrix; Independent component analysis; Laboratories; Signal processing algorithms; Software algorithms; Vectors; FPGA; ICA; IP core; algorithms;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Natural Computation, 2009. ICNC '09. Fifth International Conference on
Conference_Location :
Tianjin
Print_ISBN :
978-0-7695-3736-8
Type :
conf
DOI :
10.1109/ICNC.2009.644
Filename :
5362714
Link To Document :
بازگشت