# from state space to arma

Started by May 2, 2008
Hi all,
I would be greatfull if someone could advice me on the following problem.
I have a state space model with matrices A,B,C (x(k+1)=Ax(k)+Bu(k),state
eq. y(k)=Cx(k)) how can from this state space form to pass in a form of
arma.Can i use rls to estimate arma coefs?
"ekomninos" <ekomninos@gmail.com> writes:

> Hi all, > I would be greatfull if someone could advice me on the following problem. > I have a state space model with matrices A,B,C (x(k+1)=Ax(k)+Bu(k),state > eq. y(k)=Cx(k)) how can from this state space form to pass in a form of > arma.Can i use rls to estimate arma coefs?
See [brogan] chapter 15. --Randy @book{brogan, title = "Modern Control Theory", author = "William L. Brogan", publisher = "Prentice-Hall", edition = "second", year = "1985"} -- % Randy Yates % "Ticket to the moon, flight leaves here today %% Fuquay-Varina, NC % from Satellite 2" %%% 919-577-9882 % 'Ticket To The Moon' %%%% <yates@ieee.org> % *Time*, Electric Light Orchestra http://www.digitalsignallabs.com
On May 2, 9:54 pm, "ekomninos" <ekomni...@gmail.com> wrote:
> Hi all, > I would be greatfull if someone could advice me on the following problem. > I have a state space model with matrices A,B,C (x(k+1)=Ax(k)+Bu(k),state > eq. y(k)=Cx(k)) how can from this state space form to pass in a form of > arma.Can i use rls to estimate arma coefs?
Easy peasy... Transfer function is G(z) = C(zI-A)^-1B That should give you your ARMA (transfer function) model. K.
On May 2, 2:54&#2013266080;am, "ekomninos" <ekomni...@gmail.com> wrote:
>> Can i use rls to estimate arma coefs?
Yes, that will work in some controlled applications. I would use a minimizing routine. Scilab has optim and lsqrsolve. lsqrsolve is what you want to use because it will reduce the errors between vectors of actual and estimated values very efficiently. Peter Nachtwey
and what if i had state space system of the form x(k+1)=Ax(k)+w
y(k)=Cx(k)+v with A a nonlinear state transition matrix?
how does w and v affect ar and ma coeffs?
how does w and v affect ar and ma coeffs?