Jump to content

Talk:Phase noise

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Question

[edit]

Does phase noise effect the in , so it isn't static anymore but changes a little to smaller and higher values? --Abdull 09:49, 19 August 2005 (UTC)[reply]

Yes. 67.198.37.17 (talk) 17:20, 13 April 2019 (UTC)[reply]

Arguments against proposed merger

[edit]
  • I found the article on dBm to be very useful, with some extension the dBc article could be equally useful.
  • I would say there is a clear distinction between the unit dBc and the phenomenon of phase noise.
  • Given the lack of information in the phase noise article, I would say there isn't even short-term benefit to the merger.DJIndica 22:12, 23 March 2006 (UTC)[reply]

Example conversion to degrees

[edit]

Article says "For example, the phase noise may be -40 dBc integrated over the range of 1 kHz to 100 kHz. This Integrated phase noise (expressed in degrees) can be converted to jitter (expressed in seconds) using the following formula. Jitter(seconds) = PhaseError(degrees) / (360xFrequency(hertz))"

but could it be made clearer how to convert the integrated phase noise from dBc to degrees. Rod57 (talk) 15:22, 22 October 2008 (UTC)[reply]

The answer may be in here [1] Rod57 (talk) 22:32, 22 October 2008 (UTC)[reply]

PSD

[edit]

Is PSD Power Spectral Density? Expanding acronyms seems helpful. --Denimskater (talk) 22:22, 31 October 2017 (UTC)[reply]

Dimensions

[edit]

I changed the dimension of difference.

The article talked about things like 6 dB/Hz per octave. This should be 6 dB / octave. When talking about ratios, only dB remains.

eg.: The difference (ratio) between 30 dBm and 10 dBm is 20 dB (and not 20 dBm)

similar with: the difference ratio between -100 dBc/Hz and -110dBc/Hz equals 10 dB. 213.93.92.230 (talk) 20:32, 24 November 2008 (UTC)[reply]

I think in some contexts that the above can be incorrect: take "–20 dBc/decade slope". As I ineterpret this, it means the total power in each decade reduces by 20dB for each decade of frequency increase. That is the same as the value of L(f) reducing by 30dbc/Hz. This should definitely read "-30dBc/Hz per decade of frequency" if it is to align withthe mathematics.
In fact, I think the basic principle is wrong. If the unit is dBc/Hz the difference is also in dBc/Hz. If the difference is in dBc/octave the difference is in dBc/octave, etc.
I suspect the problem originates with the abbreviation dB/bandwidth. It means power measured in dB normalised to a signal of 1-Hz bandwidth. The dB do not scale with bandwidth (however defined).
Can the page owner please correct this?PhysicistQuery (talk) 21:44, 13 July 2016 (UTC)[reply]
The slope refers to the fall of power density; it is not about power in a decade. Glrx (talk) 22:08, 13 July 2016 (UTC)[reply]
The -20dB/decade is the standard Leeson's equation result, and nothing more. 67.198.37.17 (talk) 17:23, 13 April 2019 (UTC)[reply]

The Of

[edit]

 Done

I think there should be a word between the and of in "is one half the of the double-sideband spectral density of phase fluctuation". Any suggestions? ϢereSpielChequers 08:25, 21 April 2016 (UTC)[reply]

Thanks for catching that. As seen in a draft of IEEE STD 1139, the phrase should be "one half of the double-sideband spectral density of phase fluctuations", so there was an extra "the" and fluctuation should be plural--in phase noise, there is never just one fluctuation. I've made the change. --Mark viking (talk) 09:06, 21 April 2016 (UTC)[reply]
Thanks Mark, much appreciated. ϢereSpielChequers 16:29, 21 April 2016 (UTC)[reply]

ℒ(f)=Sφ(f)/2

[edit]

According to the definitions, ℒ(f) is the spectral content (relative to the carrier) of pure phase noise in a single sideband. Sφ(f) is the phase error relative to signal amplitude when measured on a phase detector. It seems to me that the two ℒ(f) sidebands add coherently to create Sφ(f), so the difference should be 6-dB. If this is correct, and we are expressing ℒ(f) and Sφ(f) as power relative to carrier we should have ℒ(f)=Sφ(f)/4. Of course we often see both these measures these expressed in terms of relative amplitude, so I'm not clear which of the various options I am misinterpreting. Can someone provide some clarity here? ThanksPhysicistQuery (talk) 21:57, 13 July 2016 (UTC)[reply]

Who says the noise in one sideband is coherent with the noise in the other sideband? Glrx (talk) 22:10, 13 July 2016 (UTC)[reply]
The general issue is that this article is very nearly incoherent, with no continuity from one equation to the next, no definitions of terms, no derivations, no provable or proven claims. Its sort of a concept-salad. A better version of this article might (for example?) start with φ(t) some random walk with some power spectrum and then properly define ℒ(f) and Sφ(f) and show what happens to these, in each side-band. Or conversely, start with noise in the sidebands and derive phase noise. Or something, anything would be better than what we have now. Seem's like Leeson's equation is some kind of start for actual oscillators, so the question is, how to back away from Leeson's equation and Allan variance and get to some clear, generic, understandable statements? 67.198.37.17 (talk) 17:36, 13 April 2019 (UTC)[reply]
The articles on flicker noise and oscillator phase noise are in even worse shape. Oi. 67.198.37.17 (talk) 18:00, 13 April 2019 (UTC)[reply]

Merger proposal

[edit]

I propose that Oscillator phase noise be merged to this page. The content in Oscillator phase noise page is not much and can be easily merged to this page. It would also make it more coherent by putting things in one place. pony in a strange land (talk) 07:30, 20 October 2016 (UTC)[reply]

  • Oppose. Phase noise is the general topic. The article can cover general topics such as measuring it an equivalent representations. Oscillator phase noise is a specific topic that can touch on design issues driven by the more general topic. Burn power. Device selection. Choose low NF at operating band. DC mixing means 1/f noise of the active device is significant. Resonator selection. Increasing resonator Q decreases phase noise. Oscillator phase noise article can address problems brought on by VCO requirements; FM is PM. PN requirements restrict VCO range. Band-switching the resonator. Averaging control voltage noise with multiple varactors. Both articles can have lot of meat. Glrx (talk) 22:23, 25 October 2016 (UTC)[reply]
Closing, given uncontested opposition and no support. Klbrain (talk) 09:55, 20 May 2018 (UTC)[reply]

One-Sided vs Two-Sided

[edit]

"The IEEE defines phase noise as ℒ(f) = Sφ(f)/2 where the "phase instability" Sφ(f) is the one-sided spectral density of a signal's phase deviation.[4] Although Sφ(f) is a one-sided function, it represents "the double-sideband spectral density of phase fluctuation".[5][clarification needed] " Is it one-sided or two-sided then? 2601:646:8300:10:A548:E8D8:258F:12B2 (talk) 00:05, 19 December 2020 (UTC)[reply]