Title :
Choice of weighting for averaged non-parametric transfer function estimates
Author_Institution :
Control Syst. Centre, UMlST, Manchester, UK
Abstract :
We consider the statistical properties of nonparametric transfer function estimates formed via the weighted sum of empirical transfer function estimates. In the case that the system input is considered deterministic, we show that the weighting corresponding to the classical averaged periodogram estimate is optimal. A similar asymptotic result is derived for the case where the input is considered stochastic.
Keywords :
stochastic processes; transfer functions; averaged periodogram estimate; stochastic input; transfer function estimates; Control systems; Discrete Fourier transforms; Equations; Fourier transforms; Frequency; Gaussian noise; Heat transfer; Stochastic processes; Temperature control; Transfer functions;
Conference_Titel :
Decision and Control, 2004. CDC. 43rd IEEE Conference on
Conference_Location :
Nassau
Print_ISBN :
0-7803-8682-5
DOI :
10.1109/CDC.2004.1428777
