Frequency domain identification of ARX models in the presence of additive input-output noise

This paper describes a new approach for identifying ARX models from a finite number of measurements, in presence of additive and uncorrelated white noise. The proposed algorithm is based on some theoretical results concerning the so–called dynamic Frisch Scheme. As a major novelty, the proposed approach deals with frequency domain data. In some aspects, the method resembles the characteristics of other identification algorithms, originally developed in the time domain. The proposed method is compared with other techniques by means of Monte Carlo simulations. The benefits of filtering the data and using only part of the frequency domain is highlighted by means of a numerical example.