On recursive Markov parameters estimation for MIMO systems

Abstract This work develops a recursive algorithm to estimate given size sequence of Markov parameters for linear discrete-time systems, which is related FIR models estimation. The discussion on in identification literature tends be brief due its poor prediction error low order models, although parameter shorter length can used, e.g., as the input data-driven MPC based and system combined with realization theory. Estimation larger also used applications itself not relevant, such stability assessment or norm computations. formulation derived SISO systems then we extended it MIMO case. An analysis overall truncation bias errors developed illustrative examples are highlight method’s performance. In further illustrate difference estimation results different inputs, since choice affected by method utilised.