Authors:
Jens G. Linden
;
Tomasz Larkowski
and
Keith J. Burnham
Affiliation:
Coventry University, United Kingdom
Keyword(s):
System identification, Errors-in-variables, Recursive identification.
Related
Ontology
Subjects/Areas/Topics:
Adaptive Signal Processing and Control
;
Informatics in Control, Automation and Robotics
;
Optimization Problems in Signal Processing
;
Real-Time Systems Control
;
Signal Processing, Sensors, Systems Modeling and Control
;
System Identification
;
Time Series and System Modeling
Abstract:
The implementation of a recursive algorithm for the estimation of parameters of a linear single-input singleoutput errors-in-variables system is re-considered. The objective is to reduce the computational complexity in order to reduce the computation time per recursion, which, in turn, will allow a wider applicability of the recursive algorithm. The technique of stationary iterative methods for least squares is utilised, in order to reduce the complexity from cubic to quadratic order with respect to the model parameters to be estimated. A numerical simulation underpins the theoretically obtained results.