DocumentCode
3252237
Title
A modified natural gradient algorithm for blind source separation using generalized constant modulus
Author
Fanglin Gu ; Hang Zhang ; Jiang Zhang ; Desheng Zhu ; Pan Luo
Author_Institution
Coll. of Commun. Eng., PLA Univ. of Sci. & Technol., Nanjing, China
fYear
2013
fDate
2-4 July 2013
Firstpage
663
Lastpage
666
Abstract
This paper proposes a modified natural gradient algorithm (M-NGA) for blind separation of communication signals. In the proposed algorithm, a new cost function is introduced by exploiting both the generalized constant modulus property and the statistical independent property of source signals. Then, the M-NGA is derived by maximizing the new cost function. Simulation results show that the M-NGA has faster convergence than the conventional NGA.
Keywords
blind source separation; gradient methods; M-NGA; blind source separation; communication signals; cost function; generalized constant modulus; modified natural gradient algorithm; source signals; Approximation algorithms; Convergence; Cost function; Performance analysis; Phase shift keying; Signal processing algorithms; Source separation; Blind source separation; generalized constant modulus; natural gradient;
fLanguage
English
Publisher
ieee
Conference_Titel
Telecommunications and Signal Processing (TSP), 2013 36th International Conference on
Conference_Location
Rome
Print_ISBN
978-1-4799-0402-0
Type
conf
DOI
10.1109/TSP.2013.6614019
Filename
6614019
Link To Document