Identification of multivariable systems based on finite impulse response models with flexible orders

For multi-input, multi-output stochastic systems, by means of auxiliary models - finite impulse response (FIR) models, we develop an identification algorithm to estimate the FIR model parameters of each entry (sub-submodel) of transfer matrices with an increasing order for the FIR model. The basic idea is to use auxiliary models to predict/estimate the outputs of the sub-submodels, and further to use the recursive least squares algorithm or the Pade approximation method to produce the parameter estimates of sub-submodels. Some simulation results are included.