Least squares based iterative parameter estimation algorithms for multivariate autoregressive moving average systems using the decomposition

摘要

This paper focuses on the parameter estimation problem of multivariate autoregressive moving average systems and develops a decomposition based least squares iterative identification algorithm using the data filtering. The basic idea is to transform the original system to a hierarchical identification model to decompose the hierarchical model into three subsystems and to identify each subsystem one by one. Compared with the least squares based iterative algorithm, the proposed decomposition algorithm requires less computational efforts. A simulation example is provided to test the proposed algorithm.