Prev: zero lag Butterworth filter
Next: SNR calculation using FFT
From: sw3Quant on 25 Jan 2010 17:32
hi. Iam new to the signal processing toolbox. Trying to do something pretty
simple, but having spent all day hunting around on the web and mathworks
cant find the answer.

I have a function, which acts to filter a timeseries. I would like to plot
its magnitude response using freqz.

The function is of the form:

function signal = myFunc(ts1, ts2, conf)

%ts1 and ts2 are vectors
%conf is an struct passed in containing all the parameter values for the
filter

signal = (aconf.1*ts1 + conf.a2*ts1 + conf.a3* ts1) * (conf.b1*ts2 +
conf.b2)

end

Can anyone help me understand what a call to the function looks like??

Many Thanks!!!




From: Michael Plante on 26 Jan 2010 10:27
>hi. Iam new to the signal processing toolbox. Trying to do something
pretty
>simple, but having spent all day hunting around on the web and mathworks
>cant find the answer.
>
>I have a function, which acts to filter a timeseries. I would like to
plot
>its magnitude response using freqz.
>
>The function is of the form:
>
>function signal = myFunc(ts1, ts2, conf)
>
>%ts1 and ts2 are vectors
>%conf is an struct passed in containing all the parameter values for the
>filter
>
>signal = (aconf.1*ts1 + conf.a2*ts1 + conf.a3* ts1) * (conf.b1*ts2 +
>conf.b2)
>
>end
>
>Can anyone help me understand what a call to the function looks like??

You need to give it a discrete (z-domain) transfer function. What you
have there, if I understand your notation, is time domain, and apparently a
nonlinear function of two inputs. If that's correct, transfer functions
may not be of much help (nonlinear extensions to the concept do exist, but
I'm not familiar with them).

Out of curiosity, did you mean ".*" (element mult)? Otherwise, you really
need to specify in your comments which one is a row and which one is a
column, and, preferably, enforce it.

 | 
Pages: 1
Prev: zero lag Butterworth filter
Next: SNR calculation using FFT