A new method for identifying errors-in-variables system models
We study the problem of system identification for the errors-in-variables (EIV) model, based on noisy input and output measurement data, without assuming equal noise variances at the system input and output. A new method is proposed and a new identification algorithm is developed based on the subspace interpretation. The algorithm is unbiased, and achieves the consistency of the parameter estimates. The performance of the identification algorithm is examined by a numerical example, and compared with that of the Frisch scheme based algorithm.
Publication Source (Journal or Book title)
Chinese Control Conference, CCC
Kang, H., Gu, G., & Zheng, W. (2019). A new method for identifying errors-in-variables system models. Chinese Control Conference, CCC, 2019-July, 1537-1542. https://doi.org/10.23919/ChiCC.2019.8865846