AR model parametrelerini ve derecesini tahmin etme metodları

AR model parametrelerini tahmin etmek için, en küçük kareler, ileri-geri tahmin, Burg ve geometrik kafes yöntemleri gibi dört metod sunulmuştur. AR sistem kimliklendirmede bu dört metodun performansını göstermek için, simulasyon sonuçları verilmiştir. AR model derecesi, Akaike bilgi kriteri, final tahmin hatası, minimum tanımlayıcı uzunluk ve Rissanen kriterleri gibi farklı kriterler kullanılarak belirlenmiştir. AR model derecesinin belirlenmesinde en iyi sonucun, minimum tanımlayıcı uzunluk kriteri ile elde edildiği gösterilmiştir.

Methods for estimation of AR model parameters and orderx

The four methods, the least squares, the forward-backward prediction, the Burg, and the geometric lattice, are presented for the estimation of AR (Autoregressive) model parameters. Simulation results are given to show the performances of these four methods for AR system identification. AR model order is determined by using different criteria such as the Akaike information criterion, the final prediction error, the minimum description length, and the Rissanen criterion. It was shown that the best result for the determination of the AR model order is obtained by the minimum description length criterion.

___

1. Makhoul, J., Linear Prediction: A Tutorial Review, Proc. IEEE, 63, 561-580, 1975.
2. Kay, S.M., Modern Spectral Estimation: Theory and Applications, Prentice-Hall, Englewood Cliffs, NJ, 1988.
3. Söderström, T. and Stoica, P., System Identification, Prentice Hall, Cambridge, 1989.
4. Oppenheim, A.V. and S chafer, R.W., Discrete Time Signal Processing, Prentice Hall, USA, 1989.
5. Proakis, J.G. and Manolakis, D.G., Digital Signal Processing, Prentice Hall, New Jersey, 1996.
6. Cowan, C.F.N. and Grant, P.M., Adaptive Filters, Englewood Cliffs, NJ, Prentice-Hall, 1985.
7. Burg, J.P., A New Analysis Technique for Time Series Data, NATO Advanced Study Institute on Signal Processing with Emphasis on Underwater Acoustics, August 12-23, 1968.
8. Akaike, H., A New Look at the Statistical Model Identification, IEEE Trans. Automat. Contr., AC-19, 716-723, 1974.
9. Rissanen, J., Modelling by Shortest Data Description, Automatica, 14, 465-471, 1978
10. Schwarz, G., Estimating the Dimension of a Model, Ann. Stat., 6, 461-464, 1978.
11. Priestley, M.B., Spectral Analysis and Time Series, Academic Press, 1981.
12. Akaike, H., A Bayesian Extension of the Minimum AIC Procedure of Autoregressive Model Fitting, Biometrika, 66, 237-242, 1979.
13. Rissanen, J., Stochastic Complexity and Modelling, Ann. Stat, 14, 1080-1100, 1986.
14. Wax, M., Order Selection for AR Models by Predictive Least Squares, IEEE Trans. Acoust, Speech, Signal Processing, 36, 581-588, 1988.
15. Singer, A.C. and Feder, M., Universal Linear Prediction by Model Order Weighting, IEEE Trans. Signal Processing, 47, 2685-2699, 1999.