An order-recursive algorithm for estimating pole-zero models

This paper deals with the pole-zero estimation of a discrete-time linear system from a measured input-output record. It is shown that the minimization of the squared equation error for a recursive (n, n - 1) filter can be implemented by an order-recursive algorithm using scalar products of records derived from the data. The algorithm is based on the Gram-Schmidt orthogonalization of an intertwined Krylov sequence which consists of successively time-shifted or circularly time-shifted versions of the input and output records.