The asymptotic method for the identification of errors-in-variables systems

This work studies the identification of errors-in-variables (EIV) systems. An asymptotic method (ASYM) is developed for the EIV system. The ASYM is to start with a high order model estimation and then to perform a frequency domain weighted model reduction. Firstly when applying the high order ARX (autoregressive with external input) model estimation procedure, a consistent ARX model estimation method of EIV systems is applied. The asymptotic variance expression of the estimated high order ARX model is derivated which forms the basis of the ASYM. Simulations are used to illustrate the performance of the ASYM for EIV systems.