L2-optimal identification of MIMO errors-in-variables models

This paper extends an L 2 -optimal identification method for SISO errors-in-variables models (EIVMs) to cope with a MIMO case. According to the orthogonal decomposition signals for measurements, L 2 -optimal approximate models are built, in which the system model is described by a normalized right g...

Full description

Saved in:
Bibliographic Details
Main Authors: Li-Hui Geng, Li-Yan Geng
Format: Conference Proceeding
Language:English
Subjects:
errors-in-variables
Frequency domain analysis
Linear matrix inequalities
MIMO
Noise
Numerical models
Optimization
v-gap metric
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper extends an L 2 -optimal identification method for SISO errors-in-variables models (EIVMs) to cope with a MIMO case. According to the orthogonal decomposition signals for measurements, L 2 -optimal approximate models are built, in which the system model is described by a normalized right graph symbol (NRGS) and the associated noise model (NM) by its complementary inner factor (OF). The v-gap metric is employed to optimize the system model parameters and thus the minimization problem can be solved by linear matrix inequalities (LMIs). With the estimated system model, the NM can then be obtained from a model transform. Finally, a numerical simulation is given to verify the proposed method.
DOI:10.1109/ICECENG.2011.6057287