IDENTIFICATION OF LINEAR STOCHASTIC-SYSTEMS THROUGH PROJECTION FILTERS

A novel method is presented for identifying a state-space model and a state estimator for linear stochastic systems from input and output da ta. The method is primarily based on the relationship between the stat e-space model and the finite difference mode) of linear stochastic sys tems derived through projection filters. It is proved that least-squar es identification of a finite difference model converges to the model derived from the projection filters. System pulse response samples are computed from the coefficients of the finite difference model. In est imating the corresponding state estimator gain, a z-domain method is u sed. First the deterministic component of the output is subtracted out , and then the state estimator gain is obtained by whitening the remai ning signal. An experimental example is used to illustrate the feasibi lity of the method.