MARKOV PARAMETERS CALCULATION FROM INPUT/OUTPUT DATA OF MIMO DISCRETE-TIME SYSTEMS

摘要

The paper introduces an algorithm for the direct determination of Markov parameters from samples of input/output data of linear MIMO discrete-time systems. In this sense, the algorithm may be considered as an initial step in identifying discrete- or continuous-time models of MIMO systems. In the presence of noise, the procedure behaves similarly as Least Square identification procedures. Computational examples illustrating the feasibility of the proposed algorithm are included.