Performance analysis of the recursive extended least squares method for multivariable systems with moving average noises

This paper considers the parameter estimation problems of the multivariable linear systems disturbed by moving average noises, which can be modeled by a multiple pseudolinear regression model. The recursive extended least squares algorithm is presented for this class of multivariable systems, and then the performance of the proposed algorithm is studied. The analysis shows that the parameter estimates converge fast to their true values under weak conditions. Two simulation examples are given to illustrate the effectiveness of the algorithm.