The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> 1/f noise- the DC value
https://designers-guide.org/forum/YaBB.pl?num=1363865921

Message started by iVenky on Mar 21^st, 2013, 4:38am

Title: 1/f noise- the DC value
Post by iVenky on Mar 21^st, 2013, 4:38am

I know that this question has been answered several times. Anyway I need some clear explanation on this. As we all know 1/f noise has more value near the DC. So obviously if you average the signal for a long time you are probably going to get some high value (at least little bit high). Does this mean that when you see a signal its DC (which is obtained by adding all values) is constantly increasing as the time progresses. This should mean that you don't have a stationary process (or in other words constant mean). I tried generating a 1/f noise using MATLAB (got the correct spectrum for 1/f noise) and I noticed this- When I summed all the data values for a some range (starting from 0) I got a non zero value. After this I increased the range and I got some other non-zero value (more than what I got before). So it should mean that its not a stationary process (and obviously as you keep on increasing the range the value keeps on increasing which indicates that at DC you should get infinity right?).

Title: Re: 1/f noise- the DC value
Post by raja.cedt on Mar 21^st, 2013, 7:20am

hello,
yes flicker noise has "infinite" spectral density at dc in theory, but think about dc will not be possible in the practical scenario, just because you might want to use system for some time to some time. Let's say you wanted to use for 1 year, then your lower integration limit will be 1/year in the rms power integration. once i have calculated noise power for 1sec and 1 week the difference is very less 27%, so though it looks like huge number at dc but based on the rms you get some finite number.

Unfortunately i donno creating in matlab.

Thanks,
Raj.

Title: Re: 1/f noise- the DC value
Post by iVenky on Mar 21^st, 2013, 10:40pm

raja.cedt wrote on Mar 21^st, 2013, 7:20am:

What I had done in matlab was this- I created the spectrum of 1/f noise first and then found out the ifft and created the 1/f noise. So it means that 1/f noise is somewhat related to aging of devices (which would obviously result in change in its characteristics). I read somewhere that 1/f noise is present everywhere. Do they mean the aging of the devices?

Title: Re: 1/f noise- the DC value
Post by raja.cedt on Mar 22^nd, 2013, 12:57am

hello,
i don't think it's because of the aging, for example if you take a BJT(ideally it has no flicker noise) and let it run for many years still you don't see any flicker noise means for me it looks like aging is not the cause. It just because trap's in the silicon oxide interface or some other material issue.

May be you can forward the doc where did you find this aging so that i can read learn better.

Thanks,
Raj.

Title: Re: 1/f noise- the DC value
Post by tzg6sa on Mar 23^rd, 2013, 11:40am

1/f noise has nothing to do with aging. It is due to the "traps" where the electrons run from the source to the drain. MOS devices conducts near the boundary of two material: silicon and oxide, which means plenty of irregularity in the crystalline structure. Therefore the MOS devices are more prone to 1/f noise compared to the bipolars where the electrons travels through the device within a mainly homogeneous crystalline structure. Imagine the traps as small half-sphere pits and the electrons as small balls. If the balls are fast then they will less likely to stuck in the pits. Since they have high momentum, they can get out of the pit. It's easy to see that the principle is the same if they are not sphere-like in shape. This phenomenon is present at different segments of physics and it is called generally 1/v law (I am not sure about the exact english term), where v denotes velocity. On another well know example is the radioactive decay in the nuclear plants. A so called moderator is used to slow down the electrons to spend more time near the nucleus, which increases the probability of the decay.