Identification of dynamic errors-in-variables systems with periodic data
Abstract
Using instrumental variable methods to estimate the parameters ofdynamic errors-in-variables systems with a periodic input signal isthe focus in this paper. How to choose suitable instrumentalvariable vectors is the key point. Two variants are proposed; bothof them can generate consistent estimates. An analysis shows thatthe best accuracy is achieved by using a specific overdeterminedinstrumental variable vector. Numerical illustrations demonstratethe effectiveness of the proposed Extended IV method for both white andcolored measurement noise. It is superior to alternative methodsunder low signal to noise ratios.
