Title :
Blind source separation by a geometrical method
Author :
Shindo, Hiroyuki ; Hirai, Yuzo
Author_Institution :
Master´´s Program in Sci. & Eng., Univ. of Tsukuba, Japan
fDate :
6/24/1905 12:00:00 AM
Abstract :
This paper presents an algorithm for blind source separation (BSS), which can find source orientations in the joint distribution of observed mixtures by Hough transformation and the sources are separated by projecting the mixtures onto those axes. Since the algorithm involves no iteration, it can follow the changes of source directions
Keywords :
Hough transforms; geometry; signal processing; BSS; Hough transformation; blind source separation; geometrical method; source orientations; Algorithm design and analysis; Blind source separation; Costs; Decorrelation; Independent component analysis; Neural networks; Probability distribution; Robustness; Source separation;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007649
