Abstract
In this paper, a fast algorithm for vector autoregressivemoving-average (ARMA) parameter estimation under noise environments is proposed. Based on an equivalent AR parameter model technique and a Yule-Walker equation technique, solving the parameter estimation problem of the VARMA model is well converted into solving linear equations. Therefore, the proposed algorithm has a lower computational complexity and a faster speed than conventional algorithms. Application examples with application to Lorenz systems confirm that the proposed algorithm can obtain a good solution.
Publisher
Trans Tech Publications, Ltd.
Reference16 articles.
1. P. Stoica and R. Moses, Introduction to spectral Analysis., Englewood Cliffs, NJ: Prentice-Hall, (1997).
2. S.M. Kay, Modern Spectral Estimation, Englewood Cliffs, NJ: Prentice-Hall, (1998).
3. Sheng Lu, Ki Hwan Jus, A New Algorithm for Linear and Nonlinear ARMA Model Parameter Estimation Using Affine Geometry., IEEE Transactions On Biomedical Engineering, vol. 48, no. 10, (2001).
4. J. L. Rojo-Alvarez, M. Martinez-Ramon, Support vector method for robust ARMA system identification., IEEE Transactions On Signal Processing}, vol. 52, 2004, pp.155-164.
5. G.B. Giannakis,Y. Inouye, and J.M. Mendel, Cumulant based parameter estimation of multichannel moving-average processesm, IEEE Trans Automat. Contr, pp.783-787, (1989).