|
OPTIMAL IV IDENTIFICATION AND ESTIMATION OF CONTINUOUS-TIME TF MODELS
Peter C. Young
Systems and Control Group, Centre for Research on Environmental
Systems and Statistics, Lancaster University, Lancaster LA1 4YQ, U.K.,
and
CRES, Australian National University, Canberra, Australia
The paper outlines an optimal, maximum likelihood formulation of the continuous-time transfer function estimation problem and shows how the Refined Instrumental Variable (RIVC) algorithm provides an iterative solution to this problem. With the help of a simulation example, it then compares and contrasts the performance of the RIVC algorithm with that of two other, sub-optimal, continuous-time transfer function estimation procedures that have been suggested more recently.
Keywords: continuous-time, transfer function, optimal instrumental variables, maximum likelihood, optimal estimation, comparative study
Session slot T-Tu-E02: Continuous-time model identification: methods and applications/Area code 3a : Modelling, Identification and Signal Processing
 |
