A Parameter Identification Method for Single-Input and Single-Output Linear Systems with Colored Noises Using Noise Gain Function

Abstract

A new identification method for single-input and single-output linear systems with colored input and output noises is proposed. Variances and autocovariances of the input noise are assumed to be known and those of the output noise are assumed to be unknown. The proposed method uses the eigenvector method and the noise gain functions which are derived in this paper. By using the noise gain functions, the variances and autocovariances of the output noise generated by the input noise can be obtained from those of the input noise. This algorithm uses an iterative calculation and it consists of two parts. One part is the identification of the system parameters using eigenvector. The other part is the estimation of the variance and autocovariace of output noises using the noise gain function. Some results of the simulation show the effectiveness of the proposed method.