Title :
Fast hierarchical least mean square algorithm
Author_Institution :
Dept. of Inf. Manage., Nat. Defense Manage. Coll., Taipei, Taiwan
Abstract :
We propose a hierarchical least mean square (LMS) algorithm where the taps of a filter are organized into a hierarchy, and the minimization process is performed repeatedly from the bottom level to the top level. The results of performance evaluation indicate that the proposed hierarchical LMS algorithm can speed up convergence rate and reduce the excess mean squared error (MSE) of the standard LMS algorithm.
Keywords :
FIR filters; convergence of numerical methods; least mean squares methods; LMS algorithm; MSE; convergence rate; fast hierarchical least mean square algorithm; filter taps; mean squared error; minimization process; terms-hierarchical filtering; Adaptive arrays; Convergence; Equalizers; Filtering; Filters; Helium; Intersymbol interference; Least mean square algorithms; Least squares approximation; Minimization methods;
Journal_Title :
Signal Processing Letters, IEEE
