Tracking analysis of an ARMA parameter estimation algorithm

The problem of adaptively estimating parameters of a time-varying autoregressive moving-average (ARMA) process using a constant-step-size Gauss-Newton algorithm is studied. Using weak convergence theory and the concept of prescaling, it is shown that an ordinary differential equation can be used to describe the mean behavior of the adaptive filter coefficients. Computer simulations are provided to substantiate the analysis.<
>