In search of the optimal model for a process with multiple inputs

Author

Kanjilal, P.P. ; Saha, Goutam

Author_Institution

Dept. of Electron. & Electr. Commun. Eng., Indian Inst. of Technol., Kharagpur, India

fYear

1995

fDate

15-18 Feb 1995

Firstpage

29983

Lastpage

30348

Abstract

The problem of discrete-time modelling of processes with input and output data is addressed. The objective is to produce the most representative model with adequate numbers of parameters. Conventional methods involve a computationally inefficient method of selection through exhaustive trials of all candidate models. Proposed is a new approach based on singular value decomposition (SVD), where the representative model is determined through rank-loss assessment of appropriately configured data matrix

Keywords

cardiology; physiological models; singular value decomposition; appropriately configured data matrix; computationally inefficient method; discrete-time modelling problem; heart transplant patients survival time; linear regression; linear-in-the-parameter model; multiple inputs process; optimal model; optimal variables set; rank-loss assessment; regressors; Computational modeling; Estimation error; Heart; Input variables; Joining processes; Linear regression; Matrix decomposition; Parameter estimation; Robustness; Singular value decomposition;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 1995 and 14th Conference of the Biomedical Engineering Society of India. An International Meeting, Proceedings of the First Regional Conference., IEEE

Conference_Location

New Delhi

Print_ISBN

0-7803-2711-X

Type

conf

DOI

10.1109/RCEMBS.1995.532168

Filename

532168