Prev: "Carrier trapping"?
Next: CPM
From: ikerr on 24 Apr 2010 21:47
Greetings,

I have been reading through the ITU-R BS.1770-1
(http://webs.uvigo.es/servicios/biblioteca/uit/rec/BS/R-REC-BS.1770-1-200709-I!!PDF-E.pdf)
document and have been trying to reproduce the frequency response of the
RLB weighting using the supplied filter coefficients. However, when I use
the supplied filter coefficients to construct a Bode plot in octave, the
resulting plot attenuates more quickly than the one shown in Figure 4. At
10 Hz, the attenuation is about -55 dB in my octave plot whereas the
attenuation is about -24 dB in Figure 4 of the ITU-R BS.1770-1 document.
Here is the sequence of octave instructions that I am using:

octave:84> fs = 48000;
octave:85> a = [1 -1.99004745483398, 0.99007225036621];
octave:86> b = [1 -2 1];
octave:87> sys = tf(b, a, 1 / fs);
octave:88> sysout(sys)
Input(s)
1: u_1

Output(s):
1: y_1 (discrete)

Sampling interval: 2.08333e-05
transfer function form:
1*z^2 - 2*z^1 + 1
--------------------------
1*z^2 - 1.99*z^1 + 0.99007
octave:89> bode(sys)

I'm trying to figure out what I'm doing wrong, however, thus far it is
eluding me. Can someone point me in the right direction?

Thanks,
--Ian


From: ikerr on 25 Apr 2010 10:30
>Greetings,
>
>I have been reading through the ITU-R BS.1770-1
>(http://webs.uvigo.es/servicios/biblioteca/uit/rec/BS/R-REC-BS.1770-1-200709-I!!PDF-E.pdf)
>document and have been trying to reproduce the frequency response of the
>RLB weighting using the supplied filter coefficients. However, when I
use
>the supplied filter coefficients to construct a Bode plot in octave, the
>resulting plot attenuates more quickly than the one shown in Figure 4.
At
>10 Hz, the attenuation is about -55 dB in my octave plot whereas the
>attenuation is about -24 dB in Figure 4 of the ITU-R BS.1770-1 document.
>Here is the sequence of octave instructions that I am using:
>
>octave:84> fs = 48000;
>octave:85> a = [1 -1.99004745483398, 0.99007225036621];
>octave:86> b = [1 -2 1];
>octave:87> sys = tf(b, a, 1 / fs);
>octave:88> sysout(sys)
>Input(s)
> 1: u_1
>
>Output(s):
> 1: y_1 (discrete)
>
>Sampling interval: 2.08333e-05
>transfer function form:
>1*z^2 - 2*z^1 + 1
>--------------------------
>1*z^2 - 1.99*z^1 + 0.99007
>octave:89> bode(sys)
>
>I'm trying to figure out what I'm doing wrong, however, thus far it is
>eluding me. Can someone point me in the right direction?
>
>Thanks,
>--Ian
>
>
>

Indeed it was a simple mistake. The octave bode plot's x-axis has units of
radians/second while the plot published in ITU-R BS.1770 has units of Hertz
along the x-axis. Converting to the appropriate units gave the desired
results.

--Ian
 | 
Pages: 1
Prev: "Carrier trapping"?
Next: CPM