Li Liu-Qing
Huanghuai University, Zhumadian, Henan 463000, China
Zhang Ren-Zhi
Huanghuai University, Zhumadian, Henan 463000, China
ABSTRACT
This study carried out the blind signal separation experiment based on the natural gradient algorithm for the blind mixed square wave, sine wave, amplitude modulation wave and noise signal, etc. The simulation verified that the natural gradient blind signal separation algorithm can accurate complete mixed signal separation and analysis thorouth simulating the complex random mixed source signal. Compared the blind signal separation performance of natural gradient algorithm in different step size condition. It is concluded that the selection of the step size is important to the steady-state error and the convergence speed. This study analyzed the convergence speed and the steady performance of the algorithm under the condition of fixed step size 0.005 and 0.1.
PDF References Citation
How to cite this article
Li Liu-Qing and Zhang Ren-Zhi, 2013. Research of Blind Mixed Signal Separation Technology Based on Fixed Step Size
Natural Gradient Algorithm. Information Technology Journal, 12: 3204-3208.
DOI: 10.3923/itj.2013.3204.3208
URL: https://scialert.net/abstract/?doi=itj.2013.3204.3208
DOI: 10.3923/itj.2013.3204.3208
URL: https://scialert.net/abstract/?doi=itj.2013.3204.3208
REFERENCES
- Bronkhorst, A.W., 2000. The cocktail party phenomenon: A review of research on speech intelligibility in multiple-talker conditions. Acta Acustica United Acustica, 86: 117-128.
Direct Link - Hyvarinen, A. and E. Oja, 1997. A fast fixed-point algorithm for independent component analysis. Neural Comput., 9: 1483-1492.
CrossRefDirect Link