Parameter identification using prior information

Abstract We consider the problem of identifying some unknown gain parameters in a single-input single-output transfer function, which is written as the ratio of proper, stable transfer functions. This technique is a generalization of currently available techniques of identification, which do not use prior information. It usually involves the identification of fewer parameters and is faster in convergence and less susceptible to errors caused by the presence of unmodeled dynamics. Extensions to the discrete-time systems and some multivariable systems are also covered.