Decomposition-based Gradient Estimation Algorithms for Multivariable Equation-error Systems

This paper concerns the parameter identification methods of multivariable equation-error systems. By means of the decomposition technique, the multivariable identification model is transformed into two sub-identification models and a decomposition-based stochastic gradient (D-SG) algorithm is presented for estimating the parameters of these two submodels. In order to further improve the convergence rate and the parameter estimation accuracy, we expand the innovation vectors to the innovation matrices and develop a decomposition-based multi-innovation stochastic gradient (D-MISG) algorithm. The simulation results confirm that the D-MISG algorithm can provide more accurate parameter estimates than the D-SG algorithm.