The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design Languages >> VHDL-AMS >> LNA: modeling time-domain noise
https://designers-guide.org/forum/YaBB.pl?num=1131898233

Message started by Visjnoe on Nov 13^th, 2005, 8:10am

Title: LNA: modeling time-domain noise
Post by Visjnoe on Nov 13^th, 2005, 8:10am

Hi,

I'm currently modeling an LNA (VHDL-AMS). Everything goes fine, except I don't like my modeling approach to include the effect of noise in time-domain (transient) simulations:

I use a process which generates a 'random' number (sigma equal to sqrt of integrated (white) noise), which is updated after each timestep (using hmin).

Does anyone know of a more elegant way to model this, avoiding the dependency on timestep/hmin?

Kind Regards

Title: Re: LNA: modeling time-domain noise
Post by Paul on Nov 13^th, 2005, 1:23pm

Hi,

as already mentioned in several posts in the design section, noise simulation in the time domain is arguable. Do you need this type of simulation? Could you replace it by AC-noise simulation?

Best

Paul

Title: Re: LNA: modeling time-domain noise
Post by Visjnoe on Nov 14^th, 2005, 11:40am

Hi Paul,

Yes, I could simply run an AC noise analysis on the LNA.

But I'm modeling an RF receiver signal path and I'd like to run a (behavioral) simulation of the entire signal path including effects like distortion and noise. Since this path includes circuits like mixers, I have to resort to time-domain simulation.

Furthermore, I have to say, that - when you set up the parameters right - the ELDO .noisetran analysis provides accurate results (verified this analysis with several trivial examples which could be predicted by handcalculations).

Kind Regards

Peter

Title: Re: LNA: modeling time-domain noise
Post by Paul on Nov 15^th, 2005, 12:06pm

Hi Peter,

I did not mean that ELDO wasn't accurate. I don't really believe in the efficiency of transient noise simulation (somehow you have to do statistics on the output noise to get standard deviation and so on...).

Paul