A modified bias compensation method for output error systems with colored noises

This paper studies the modelling and identification problems for output error systems. In order to obtain the unbiased recursive estimates of the systems, this paper presents a modified bias compensation based recursive least squares (BC-RLS) identification algorithm by means of the prefilter idea and bias compensation principle. Compared with the previous contributions in the bias compensation technique, the proposed algorithm simplifies the bias compensation recursive least squares method and can give the unbiased estimation of the system model parameters in the presence of colored noises, further more, and can be on line implemented. Finally, the advantages of the proposed BC-RLS algorithm over the other approaches are shown by simulation tests.