Identification of Wiener Models Using Support Vector Machine

The least squares support vector machines (LS-SVM) regression is presented for the purpose of nonlinear dynamic system identification. LS-SVM are used for system identification of Wiener models with memoryless nonlinear blocks and linear dynamical blocks. LS-SVM achieves higher generalization performance. The identification procedure is illustrated using two simulated examples. The results indicate that this approach is effective.