Iterative Extension of Prony's Method for ARMA Signal Modeling

摘要

A new iterative version of the Prony method is presented and shown to beexceptionally effective in finding ARMA models for acoustic data in the signal domain. The method is based on a quadratic type of gradient algorithm where it is shown that the gradient and Hessian are easily computed from the data. The new algorithm is found experimentally to have excellent convergence behavior. The performance of the algorithm is demonstrated and compared to that of the basic Prony method and to that of the Steiglitz and McBride iterative prefiltering algorithm on some recorded acoustic data. Prony's method, ARMA modeling.