LINEAR-SYSTEM IDENTIFICATION VIA BACKWARD-TIME OBSERVER MODELS

This paper presents an algorithm to identify a state-space model of a linear system using a backward-time approach. The procedure consists o f three basic steps. First, the Markov parameters of a backward-time o bserver are computed from experimental input-output data. Second, the backward-time observer Markov parameters are decomposed to obtain the backward-time system Markov parameters (backward-time pulse response s amples) from which a backward-time state-space model is realized using the eigensystem realization algorithm. Third, the obtained backward-t ime state-space model is converted to the usual forward-time represent ation. Stochastic properties of this approach will be discussed. Exper imental results are given to illustrate when and to what extent this c oncept works.