DocumentCode :
2109638
Title :
Blind source separation based on a novel relative gradient
Author :
Xue, Yunfeng ; Wang, Yujia ; Sun, Qiudong
Author_Institution :
Sch. of Electron. & Electr. Eng., Shanghai Second Polytech. Univ., Shanghai, China
fYear :
2010
fDate :
17-19 Dec. 2010
Firstpage :
818
Lastpage :
821
Abstract :
In this paper, a novel relative gradient is proposed to solve the blind source separation problem. An iterative method is introduced to solve the nonlinear matrix equation which is derived from the relative gradient where no learning rate is needed. Kernel density estimation is utilized to estimate the density functions as well as their first and second derivatives, which makes the algorithm adaptive to the unobserved sources. Computer experiments confirm the efficiency of the proposed method.
Keywords :
blind source separation; gradient methods; matrix algebra; blind source separation; iterative method; kernel density estimation; nonlinear matrix equation; novel relative gradient; Algorithm design and analysis; Blind source separation; Equations; Iterative methods; Kernel; Mathematical model; Signal processing algorithms; Blind source separation; Independent component analysis; relative gradient;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Theory and Information Security (ICITIS), 2010 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-6942-0
Type :
conf
DOI :
10.1109/ICITIS.2010.5689704
Filename :
5689704
Link To Document :
بازگشت