Parametrik olmayan metodların performans analizi

Bu çalışmada, kutup, varyans ve ortalama karesel hata gibi istatistiksel ölçümlerin değerlendirilmesi ile parametrik olmayan metodların performans analizi yapılmıştır. Parametrik olmayan metodlarda güç yoğunluk spektrumu ve spektral kestirimin tanımları incelenmiştir. Parametrik olmayan metodların hesaplanmasında hızlı Fourier dönüşümü kullanılmıştır. Varyans, kutup,çözünürlük, spektral sızıntı ve bozulma gibi spektral kestirim özelliklerine göre bu metodların performansları belirlenmiştir. Beyaz gürültünün, beyaz gürültü içinde bulunan sinüs ve kare dalga işaretlerinin güç yoğunluk Spektrum kestirimleri periodogram, korrelogram, Blackman-Tukey, Bartlett ve Welch metodları ile elde edilmiş ve bu metodların performanslarındaki farklılıklar açıklanmıştır.

Performance analysis of nonparametric methods

In this study, performance analysis of nonparametric methods were done by the evaluation of statistical measurements such as bias, variance and mean square error. The definitions of power spectrum density and spectral estimation were examined in nonparametric methods. Fast Fourier transform was used in the calculation of nonparametric methods. Performances of these methods were determined according to spectral estimation properties such as variance, bias, resolution, spectral leakage and smearing. Power spectrum density estimations of white noise, sine and square wave signals existing in white noise were obtained by periodogram, correlogram, Blackman-Tukey, Bartlett and Welch methods and the differences in performances of these methods were explained.

___

1. Kay, S.M., Modern Spectral Estimation: Theory and Application, Prentice-Hall, Englewood Cliffs, NJ, 1988.
2. Stoica, P., Moses R., Introduction to Spectral Analysis, Prentice-Hall, Upper Saddle River, NJ, 1997.
3. Oppenheim, A.V., Schafer, R.W., Digital Signal Processing, Prentice-Hall, Englewood Cliffs, NJ, 1975.
4. Helstrom, C.W., Elements of Signal Detection and Estimation, Prentice-Hall, Englewood Cliffs, NJ, 1995.
5. Proakis, J.G., Manolakis, D.G., Digital Signal Processing Principles, Algorithms and Applications, Prentice-Hall, Englewood Cliffs, NJ, 1996.
6. Kay, S.M., Marple, S.L., Spectrum Analysis- A Modern Perspective, Proceedings of the IEEE, vol.69, N o . l l , 1380-1419,1981.
7. Zivanovic, M., Carlosena A., Nonparametric Spectrum Interpolation Methods: A Comparative Study, IEEE Instrumentation and Measurement Technology Conference, vol.2, 618-623,2000.
8. Neira, L.R., Constantinides, A.G., Power Spectrum Estimation from Values of Noisy Autocorrelations, Signal Processing, vol.50, 223-231,1996.
9. Yin, Y.Q., Krishnaiah, P.R., On Some Nonparametric Methods for Detection of the Number of Signals, IEEE Transactions on Acoustics, Speech, and Signal Processing, vol.ASSP-35, No.11 , 1533-1538, 1987.