Type
Text
Type
Dissertation
Advisor
Rachev, Svetlozar | Mullhaupt, Andrew P | Pinezich, John | Riedel, Kurt.
Date
2014-12-01
Keywords
Applied mathematics | identification, MIMO, reduction, state-space
Department
Department of Applied Mathematics and Statistics.
Language
en_US
Source
This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.
Identifier
http://hdl.handle.net/11401/76471
Publisher
The Graduate School, Stony Brook University: Stony Brook, NY.
Format
application/pdf
Abstract
Consider a discrete-time linear time invariant (LTI) d-dimensional innova- tions model, z(t+1) = Az(t)+Bx(t), (1) y(t) = Cz(t)+x(t), (2) where y (t) is a sequence of d-dimensional measurement vectors and z (t) is a state vector of dimension n. The triangular input balanced (TIB) representation of the LTI system was introduced by A.Mullhaupt and K.Riedel [1] in 1995. In the Single-Input Single-Output (SISO) case, the TIB pair is uniquely determined by the poles of the system. However, the poles alone are not enough to fully characterize the Multi-Input Multi-Output (MIMO) TIB pair. We parametrize MIMO transfer functions in terms of data of the Schur tangential algorithm. The inner part appears as a Blaschke-Potapov factorization. We relate these parameters to TIB and lattice realizations, and use this correspondence to con- struct novel methods for model reduction and identification. | 108 pages
Recommended Citation
Yu, Xiao, "Identification and Model Reduction of MIMO systems in Triangular Input Balanced Form" (2014). Stony Brook Theses and Dissertations Collection, 2006-2020 (closed to submissions). 2389.
https://commons.library.stonybrook.edu/stony-brook-theses-and-dissertations-collection/2389