Signal processing using singular spectrum analysis for nonlinear system identification

System identification is defined as finding mathematical models of systems, using experimental measurements and observations. This paper proposes an identification approach based on the singular spectrum analysis (SSA) and least squares support vector machines (LS-SVM) model. The SSA is used in the pre-processing stage for de-noising the measurement data and then the LS-SVM model is trained by the de-noised data. The proposed approach was employed for identification of two nonlinear systems. The simulation results demonstrated the promising performance of the proposed approach and favorable capabilities of the SSA for nonlinear system identification.