On closed-loop system identification using polyspectral analysis given noisy input-output time-domain data (Q5926184)

The paper deals with an identification of a closed-loop system. In particular, it is assumed that the considered system is given by noisy input-output measurements and the transfer function depends on unknown parameters. The aim of the paper is to introduce estimates of the transfer function. To investigate the transfer function estimates, the statistical properties of the corresponding parameter estimates are investigated. At the end, the introduced approach is illustrated by a simulation example. The paper follows former results on a similar topic. Especially, a rich survey of the previous results is introduced and, moreover, the paper is completed by a list of references.