Notes on Digital Signal Processing: Practical Recipes for Design, Analysis and Implementation

Note 70. Fitting All-Pole Models to Deterministic Signals: Autocorrelation Method

This note describes the autocorrelation method for finding the parameters needed to fit an all-pole model to a finite sequence of samples obtained from a deterministic signal. Although the genesis of each method is different, the autocorrelation method is compuatationally identical to the Yule-Walker method described in Note 69.

The autocorrelation method is a technique for fitting an all-pole model to a deterministic signal that is assumed to be autoregressive, but where knowledge about the signal is limited to a sequence of N samples, x[0] through x[N –1]. This technique is based on normal equations that result from perfroming an error minimization over the semi-infinite ...