A New Variable Step Size LMS Adaptive Filtering Algorithm and Its Simulations

李方伟,张浩
2009-01-01
Abstract:Aiming at the disadvantage of the variable step size LMS adaptive filtering algorithms' convergence speed contradicting its steady-state error,a novel non-liner functional relationship between μ(n)and error signal e(n) was established.This function relationship has the property of increasing automatically at the beginning of this algorithm and when the unknown system is changing with time and being smaller during the steady state;it can also eliminate the interference of input uncorrelated noise on μ(n).On the basis of the functional relationship,a new algorithm of variable step size LMS adaptive filtering was presented.The theoretical analysis and computer simulations show that the performance of this algorithm is better than that of other algorithms.
What problem does this paper attempt to address?