A modeling method based on ML-DC algorithm for non-Gaussian colored processes

Gaussian mixture autoregressive model is usually used to fit the probability density and power spectrum density of non-Gaussian colored processes. Its parameters can be estimated through the ML-DC algorithm. After descriptions of the model and the estimation problem, maximum likelihood estimation for autoregressive parameters and the dynamic clutter algorithm for Gaussian mixture parameters are deduced, respectively. Based on these, ML-DC algorithm for coupling estimation between power spectrum density parameters and probability density parameters is built up. Finally, a numerical instance is illustrated where performance of estimation is discussed in detail.