Errors-in-variables identification of dynamic systems in general cases

This paper deals with the identification of dynamic systems from noisy input-output observations where the noise-free input is not parameterized. By extending the previous work Zhang et al. (2013), the present work aims to treat the errors-in-variables identification in more general cases, including correlated input-output noises, nonminimum phase or unstable linear dynamic system. It is shown by simulated examples that consistent identification can be obtained for the parameters of the plant model and the covariance matrix of the input-output noises in these unconventional setups.