TRANSFER-FUNCTION ESTIMATION FROM NOISY INPUT AND OUTPUT DATA

Two new types of bias-eliminated least-squares (BELS) based algorithms are proposed for consistent identification of linear systems with noi sy input and output measurements. It is shown that estimation of the n oise variances can be implemented through one-dimension over-parametri zation of the system transfer function. The two modified BELS algorith ms are attractive and meaningful in that noisy data are used directly in identification with no prefiltering and a direct estimate of system parameters is given without any parameter transformation. Simulation examples are included to demonstrate the effectiveness of the two prop osed algorithms. (C) 1998 John Wiley & Sons, Ltd.