A Performance Bound for a Cascade LMS Predictor

Abstract

The performance of an FIR cascade structure for adaptive linear prediction is studied, in which each stage FIR filter is independently adapted using LMS algorithm. In this paper, the performance bound is derived for cascade LMS predictor under some assumptions. We discover that this bound is possible to be better than that of the linear predictive coding (LPC) technique using block-based Levinson-Durbin algorithm if the cascade LMS predictor is well selected. We show some examples of this bound for synthetic and real audio signals in which a cascade LMS predictor outperforms the LPC technique using Levinson-Durbin algorithm.

High sampling rate audio signal modeling

“…audio signal modeling, the "independence assumption" is not always proper. In [9], we gave out the performance bound of LMS predictor without "independence assumption" as…”