Abstract
This paper deals with identification of multivariable fractional order systems. The state space model is used, based on the Guidorzi canonical form. An output error identification approach is used, and the parametric sensitivity functions model is developed as a multivariable fractional model. Its simulation is carried out using the fractional bounded integration operator. Different simulations test the method efficiency, and its statistical performance is analyzed, using Monte Carlo simulation at various signal-to-noise ratios.