Filtering based recursive least squares identification for non-uniformly sampled systems

In this paper, a filtering based recursive least squares algorithm is derived for identification of the non-uniformly sampled Box-Jenkins systems. The basic idea is to use an estimated noise transfer function to filter the input-ouput data, to obtain two identification models containing the parameters of the system model and the noise model respectively, and to present the filtering based recursive least squares method to identify the parameters of these two models, by replacing the unmeasurable terms in the information vectors with their estimates. Finally, an illustrative example is given to indicate that the proposed algorithm can generate more accurate parameter estimation compared with the auxiliary model based recursive generalized extended least squares algorithm.