A New Kalman Filter Based 2D AR Model Parameter Estimation Method

ABSTRACT

This paper presents a new method based on the Kalman filter (KF) for the two-dimensional (2D) autoregressive model parameters estimation. For this purpose, the corresponding relations of KF in one-dimensional (1D) case are extended to the 2D case and the related algorithm is presented. Main advantages of the proposed method are: online and recursive parameter estimation, and detection of parameters changes. Meantime, because the presented approach does not involve complex and time consuming matrix calculations, so it is computationally efficient. Numerical examples are expressed to verify the efficiency of the proposed method.

KEYWORDS:

• Autoregressive model
• Identification
• Kalman filter
• Parameter estimation
• Two dimensional
• Time series

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Mahdi Zeinali

Mahdi Zeinali received the BS degree in control engineering from the Sahand University of Technology, Tabriz, Iran, in 2001 and his MSc degree in control engineering from Sharif University of Technology, Tehran, Iran, in 2004. He is currently an assistant professor in the Department of Electrical Engineering at Sahand University of Technology, Tabriz, Iran. His interests are in the area of system identification and adaptive control.

Email: [email protected]

Masoud Shafiee

Masoud Shafiee received his BSc degree in mathematics in 1975, his MSc degree in mathematics and MSc degree in systems engineering from Wright State University, Dayton, Ohio, in 1980 and 1982, respectively, and his MSc and PhD degrees in electrical engineering from Louisiana State University, Baton Rouge, LA, in 1984 and 1987, respectively. He is currently full professor in the Department of Electrical Engineering at Amirkabir University of Technology, Tehran, Iran. He is the author of over 230 research papers and 13 books (as author) and 11 books (as translator). His research interests include multidimensional (M-D) system, singular systems, information and communications, robotics, and network stability.

Email: [email protected]