Variable Step-size CMA Blind Equalization based on Non-linear Function of Error Signal

2009 ◽  
Author(s):  
Zhang Liyi ◽  
Chen Lei ◽  
Sun Yunshan
Keyword(s):  
Linear Function ◽  
Blind Equalization ◽  
Variable Step Size ◽  
Error Signal ◽  
Step Size ◽  
Variable Step ◽  
Non Linear

A Variable Step Size Blind Equalization Algorithm Based on Gamma Distribution

2012 ◽  
Vol 457-458 ◽  
pp. 961-967
Author(s):  
Xiang Zhao ◽  
Jin Yong Sun ◽  
Han Hong Tan
Keyword(s):  
Gamma Distribution ◽  
Blind Equalization ◽  
Weight Vector ◽  
Variable Step Size ◽  
Constant Modulus Algorithm ◽  
Step Size ◽  
Equalization Algorithm ◽  
Variable Step ◽  
Mean Square Errors ◽  
Better Than

A variable step size Constant Modulus Algorithm (CMA) based on the gamma distribution is implemented as solutions to optimize the problem of blind equalization. The factor of step size in blind equalization algorithm is varied with gamma variable, in terms of the characteristics of which, the algorithm can search for the globe optimal equalizer weight vector. Simulation results indicate that the convergence rate and the steady Mean Square Errors (MSE) performances of the algorithm proposed are much better than conventional CMA and modified CMA blind equalization algorithms.

Adaptive Blind Equalization with Variable Step-Size Modified by Attenuation Function

2011 ◽  
Vol 474-476 ◽  
pp. 1792-1796
Author(s):  
Ying Xiao ◽  
Zhen Xing Li
Keyword(s):  
Convergence Rate ◽  
Adaptive Algorithm ◽  
Blind Equalization ◽  
Variable Step Size ◽  
Step Size ◽  
Attenuation Function ◽  
Variable Step ◽  
Better Than

Step-size is one of the important parameters which influence the performance of blind equalization, a variable step-size method control by attenuation function according to instantaneous gradient and iterative times is proposed in this paper. Initially, set larger step-size to obtain faster convergence, with iterative times increasing, step-size would decrease control by attenuation function for instantaneous gradient reducing. Step-size changes according to this method consistent with the demand variable step-size adaptive algorithm that can obtain combination of convergence rate and convergence precision to improve the performance of blind equalization. Simulations show that this method performs better than conventional blind equalization with constant step-size.

Sign in / Sign up

Export Citation Format

Share Document