A Novel Variable Step Size LMS Algorithm Based on Modified Hyperbolic Tangent and Its Simulation
2012 ◽
Vol 490-495
◽
pp. 1426-1430
◽
Keyword(s):
Low Snr
◽
In the paper, a new variable step size LMS algorithm based on modified hyperbolic tangent is presented. In the algorithm, the step size is adjusted by the estimation of the autocorrelation between and .The algorithm introduces the compensation monomial to improve the convergence and the parameters to improve the shape and bottom characteristic of hyperbolic tangent. Therefore, the algorithm has faster convergence, better performance of noise suppression,lower steady state error and misadjustment. The theoretical analysis and simulation results all show that the overall performance of the new algorithm exceeds greatly some existent others under low SNR condition.
2019 ◽
Vol 67
(6)
◽
pp. 405-414
◽
Keyword(s):
2013 ◽
Vol 756-759
◽
pp. 3972-3976
◽
Keyword(s):
2014 ◽
Vol 21
(6)
◽
pp. 653-657
◽
Keyword(s):
2013 ◽
Vol 475-476
◽
pp. 1060-1066
Keyword(s):
Keyword(s):
Keyword(s):
2013 ◽
Vol 373-375
◽
pp. 1159-1163
Keyword(s):
2003 ◽
Vol 16
(1)
◽
pp. 127-135
Keyword(s):
2014 ◽
Vol 631-632
◽
pp. 116-120
Keyword(s):