Re: tfestimate vs fft

[Henry Gomersall]

On Wed, 2013-04-24 at 14:03 +0100, Arnaud Miege wrote:
> On 24 April 2013 13:26, Henry Gomersall <address@hidden> wrote:
>         On Wed, 2013-04-24 at 12:31 +0100, Arnaud Miege wrote:
>         > I am trying to estimate the transfer function of a system
>         (SISO) based
>         > on time-domain data. I have tried by directly taking the FFT
>         of the
>         > time-domain data and also by using tfestimate.
>         
>         
>         What was the input to the system from which you derived the
>         time-domain
>         data?
>         
>         hen

> The input to the system is the heater input for the pin under
> closed-loop control (% of full scale) and the output is the
> temperature of that same pin. Both were logged on power-up of the
> machine, when it warmed up from ambient to operating temperature.

So what does taking the FFT of that data tell you? You're only going to
get something akin to the transfer function if your input signal is full
bandwidth.

Ideally, you want to whack an impulse into the system and see what
happens on the other side. Failing that you can get a pretty good idea
of the PSD (and hence the _amplitude_ of the FFT) using white (or
whitish) noise as the input, though that doesn't tell you much about the
phase.

I don't really know what tfestimate is doing, so I can't comment on its
reliability, though I'd be far more inclined to trust it than your
homebrew estimate, given that presumably it factors in all these issues.

Cheers,

hen