In many applications noise is clearly present. Under those circumstances, the reliability of the direct estimates of the impulse response function g(t) or frequency function G(e jω ) can be significantly reduced. Therefore, in Chap. 4 correlation methods, which are less sensitive to noise and thus very useful under practical circumstances, are presented. In particular, the so-called Wiener–Hopf relationship is derived from input–output data and analyzed with respect to its filter properties. The chapter finishes with spectral analysis methods that provide a transfer-function estimate using power spectra.
KeywordsAutocorrelation Function Spectral Analysis Method Hopf Equation White Noise Signal Frequency Domain Technique
- [BE83]D.R. Brillinger, P.R. Krishnaiah (eds.), Handbook of Statistics 3: Time Series in the Frequency Domain (North-Holland, Amsterdam, 1983) Google Scholar
- [EO68]L.D. Enochson, R.K. Otnes, Programming and analysis for digital time series data. Technical report, US Dept. of Defense, Shock and Vibration Info. Center, 1968 Google Scholar
- [Lju99b]L. Ljung, System Identification—Theory for the User, 2nd edn. (Prentice Hall, New York, 1999) Google Scholar
- [Mar87]S.L. Marple, Digital Spectral Analysis with Applications (Prentice-Hall, New York, 1987) Google Scholar
- [SOS00]L. Sun, H. Ohmori, A. Sano, Frequency domain approach to closed-loop identification based on output inter-sampling scheme, in Proceedings of the American Control Conference, vol. 3 (2000), pp. 1802–1806 Google Scholar
- [WG04]E. Wernholt, S. Gunnarsson, On the use of a multivariable frequency response estimation method for closed loop identification, in Proceedings of the IEEE Conference on Decision and Control, vol. 1 (2004), pp. 827–832 Google Scholar