The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Question about Fig. 17 of the paper "Device noise simulation of DS Modulators"
https://designers-guide.org/forum/YaBB.pl?num=1260505047

Message started by alireza on Dec 10^th, 2009, 8:17pm

Title: Question about Fig. 17 of the paper "Device noise simulation of DS Modulators"
Post by alireza on Dec 10^th, 2009, 8:17pm

Hi,

In Fig 17 of the paper: "Device noise simulation of DS Modulators", simulation results of the input referred noise PSD of a discrete-time integrator is plotted. The plot shows that at relatively high frequencies (close to fs/2), the noise floor starts to rise significantly. In the paper, it is mentioned that this is due to the small integrator gain near fs/2. However, this is not clear to me. The reason is I think the integrator gain should only affect the output noise spectrum, not the input-referred spectrum. (The input-referred spectrum should not depend on the transfer function H(z).)
On the other hand, I thought the input PSD should be white at medium to high frequencies.

Any feedback or comments are very much appreciated.

Thank you,

Alireza

Title: Re: Question about Fig. 17 of the paper "Device noise simulation of DS Modulato
Post by pancho_hideboo on Dec 11^th, 2009, 4:14am

alireza wrote on Dec 10^th, 2009, 8:17pm:

The reason is I think the integrator gain should only affect the output noise spectrum, not the input-referred spectrum.
(The input-referred spectrum should not depend on the transfer function H(z).)

Still you can't understand correctly.
Again see http://www.designers-guide.org/Forum/YaBB.pl?num=1258339986/25#25

Output noise spectrum is a result of contribution from all sidebands, that is, multiple foldings.
While H(z=e^jωTs) considers only zero sideband.
So the input-referred spectrum depend on the transfer function H(z=e^jωTs).

alireza wrote on Dec 10^th, 2009, 8:17pm:

On the other hand, I thought the input PSD should be white at medium to high frequencies.

Not correct.

Title: Re: Question about Fig. 17 of the paper "Device noise simulation of DS Modulators"
Post by alireza on Feb 19^th, 2010, 8:55am

On page 14 of the paper, it says Figs 15-17 are plotted up to the frequency fs/2=62.5KHz. However, the plots extend over 1MHz freq. range. Why is that?

What I am looking for is whether pnoise (time-domain) or pnoise (sources) were used to obtain the plots in Figs 15-17.

Regards.