Title :
A frequency response-based model order selection criterion
Author :
Cloud, David J. ; Kouvaritakis, Basil
Author_Institution :
Dept. of Eng. Sci., Oxford Univ., UK
fDate :
6/1/1988 12:00:00 AM
Abstract :
The use of weighting sequence models to describe the dynamics of physical systems provides an effective means of translating the uncertainty associated with the model parameter estimates derived from noisy input/output data into corresponding frequency response-uncertainty information. However, an appropriate truncation level must be established to accomplish this task. The truncation problem is addressed from a frequency response perspective, and a criterion based on frequency response considerations is proposed for selecting the proper truncation
Keywords :
frequency response; modelling; parameter estimation; frequency response-based model order selection criterion; frequency response-uncertainty information; modelling; parameter estimation; truncation; weighting sequence models; Clouds; Councils; Frequency estimation; Frequency response; Least squares approximation; Noise measurement; Parameter estimation; Testing; Uncertainty; White noise;
Journal_Title :
Automatic Control, IEEE Transactions on