Determination of the Two-Dimensional ARMA Model Order Using Rank Test Based Approach

Determination of the order of a two-dimensional (2-D) ARMA model is a key first step toward the goal of modeling a process. In this paper, a rank test approach for 2-D ARMA model order determination is proposed. This method shows that the information of AR and MA orders are implicitly contained in two different correlation matrices and the AR and MA orders of the 2-D ARMA model can be independently determined before parameter estimation. In the proposed method, the model is assumed causal, stable, linear, and spatial shift-invariant with P1times P2 quarter plane (QP) support. Analysis and simulation results demonstrate the applicability of this approach.

[1]  Xianda Zhang,et al.  High resolution two-dimensional ARMA spectral estimation , 1991, IEEE Trans. Signal Process..

[2]  Luis Torres,et al.  A new approach to texture coding using stochastic vector quantization , 1994, Proceedings of 1st International Conference on Image Processing.

[3]  Surendra Ranganath,et al.  Two-dimensional linear prediction models-part I: Spectral factorization and realization , 1985, IEEE Trans. Acoust. Speech Signal Process..

[4]  S. Rital,et al.  Two-dimensional non-Gaussian autoregressive model order determination , 2002, IEEE Signal Processing Letters.

[5]  Wen-Rong Wu,et al.  Image Estimation Using Fast Modified Reduced , 1992 .

[6]  P. Kiernan Two-dimensional AR spectral estimation using a two-dimensional minimum free energy method , 1995, IEEE Trans. Signal Process..

[7]  J. Woods,et al.  Estimation and identification of two dimensional images , 1981, 1981 20th IEEE Conference on Decision and Control including the Symposium on Adaptive Processes.

[8]  Petre Stoica,et al.  Two-dimensional system identification using amplitude estimation , 2002, IEEE Signal Processing Letters.

[9]  Xianda Zhang,et al.  Determination of the MA order of an ARMA process using sample correlations , 1993, IEEE Trans. Signal Process..

[10]  Will Gersch,et al.  Estimation of the autoregressive parameters of a mixed autoregressive moving-average time series , 1970 .

[11]  M. Karrari,et al.  AR order determination of a 2-D ARMA model , 2007, 2007 European Control Conference (ECC).

[12]  Brahim Aksasse,et al.  Two-dimensional autoregressive (2-D AR) model order estimation , 1999, IEEE Trans. Signal Process..

[13]  Brahim Aksasse,et al.  A rank test based approach to order estimation. I. 2-D AR models application , 1999, IEEE Trans. Signal Process..

[14]  J. Cadzow,et al.  Two-dimensional spectral estimation , 1981 .

[15]  Rangasami L. Kashyap,et al.  Image data compression using autoregressive time series models , 1979, Pattern Recognit..