•  
  •  
 

Corresponding Author

Soliman, Fayez

Subject Area

Electronics and Communication Engineering

Article Type

Original Study

Abstract

Various methods have been proposed for adapting the coefficients od digital lattice filters used in linear predictive speech coding systems. The most common of these is the stage-by-stage algorithm based on the least mean square (LMS)) METHOD. In this paper a new sequential adaptive algorithm referred to as "end-point" method is introduced. A comparison study between the stage-by-stage and the "end-point" has been carried out by computer simulation. The preliminary results showed that the "end-point" provides higher convergence rate and less misadjustment in mean-square error over the stage-by-stage method.

Share

COinS