The filtering based maximum likelihood recursive least squares estimation for multiple-input single-output systems

In this paper, we use a noise transfer function to filter the input-output data and propose a new recursive algorithm for multiple-input single-output systems under the maximum likelihood principle. The main contributions of this paper are to derive a filtering based maximum likelihood recursive least squares (F-ML-RLS) algorithm for reducing computational burden and to present two recursive least squares algorithms to show the effectiveness of the F-ML-RLS algorithm. In the end, an illustrative simulation example is provided to test the proposed algorithms and we show that the F-ML-RLS algorithm has a high computational efficiency with smaller sizes of its covariance matrices and can produce more accurate parameter estimates.