A new look at system parameter estimation via signal processing

Signal processing is used to obtain unbiased parameter estimates for stochastic systems subject to correlated disturbances. A filter is designed and inserted artificially into the identified system so that the resulting system has some known zeros which can be used to estimate the bias arising from the correlated disturbances. It is shown that the proposed algorithm gives a consistent estimate and that the application of signal processing to system identification is very heuristic