Maximum likelihood iterative identification approaches for multivariable equation-error moving average systems

This is the authors' abstract. We don't add key points for this paper.

The identification issues of a multivariable system with coloured noises are investigated in this paper. The decomposition strategy is utilised for dimension reduction by transforming a multivariable system into several sub-models. A maximum likelihood least squares-based iterative identification approach is presented to enhance the parameter estimation accuracy by combining the iterative identification technique with the maximum likelihood principle. A simulation example is offered to test the effectiveness of the proposed method.