International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 42 - Number 9 |
Year of Publication: 2012 |
Authors: Badreddine Louhichi, Ahmed Toumi |
10.5120/5723-7783 |
Badreddine Louhichi, Ahmed Toumi . Identification of MIMO Hammerstein models using Singular Value Decomposition approach. International Journal of Computer Applications. 42, 9 ( March 2012), 29-37. DOI=10.5120/5723-7783
In this paper, we present a new approach to identify multivariable Hammerstein systems based on the Singular Value Decomposition (SVD) method. The technique allows for the determination of the memoryless static nonlinearity as well as the estimation of the model parameters of the dynamic Auto-Regressive model with eXogenous input (ARX) part. First of all, an iteration procedure is proposed to identify the parameters of Multi-Input Multi-Output (MIMO) Hammerstein models by using the Recursive Least Squares (RLS) algorithm. Secondly, the obtained parameter estimates of the identification model include the product terms of the parameters of the original systems. So, to separate these parameters of the original parameters from the product terms, the singular value decomposition method is discussed. Finally, a simulation study is performed to demonstrate the effectiveness of the proposed method compared with the existing approaches.