Bias Compensated Weighted Least Squares Method for Multiple-Input Single-Output Errors-In-Variables Models Identification

Abstract

This paper addresses the identification of multiple-input single-output (MISO) errors-in-variables (EIV) models, in which all input and output measurements are corrupted by white noise with unknown variances. To achieve consistent parameter estimation while reducing the need to estimate all unknown noise variances, a bias compensation principle based method called the bias compensated weighted least squares (BCWLS) method was previously proposed for singleinput single-output (SISO) EIV models. The present paper extends the BCWLS method, originally developed for the SISO case, to the MISO case. Simulation results demonstrate the effectiveness of the proposed recursive algorithm.