[SciPy-user] Fwd: signal.lti(A,B,C,D) with D=0

josef.pktd@gmai... josef.pktd@gmai...
Tue Feb 17 15:09:32 CST 2009


On Tue, Feb 17, 2009 at 3:25 PM, flyeng4 <williamhpurcell@gmail.com> wrote:
> I am getting back to using signal.lti for state-space representation.
> This is a post and a reply that I had received a while back, but there
> was no follow up. Any thoughts on a solution for representing MIMO or
> SIMO systems with signal.lti, with the issue described below with
> ss2zpk?
> -Bill


I was looking recently into scipy.signal to see whether it can be used
for time series analysis.

After digging around in the code and trying out some examples (I
didn't find any examples in the docs), I got the impression that
scipy.signal is not designed to handle more than one signal (I wanted
recursive filter with interactions between several time series). I
think the main problem I found was, that numpy/scipy doesn't have a
multivariate/multidimensional polynomial class. Somewhere when
converting from a multidimensional transfer function, scipy signal
uses numpy.polynomial which only works for one-dimensional
polynomials.
That's when I gave up.

My conclusion was, that for multidimensional signals, scipy signal
would need to be rewritten quite a bit.

Josef


More information about the SciPy-user mailing list