Eric Jacobsen <eric.jacobsen@ieee.org> writes:> [...] > This place is going to get really tiresome if every response to a > question has to include every possible nuance in the answer.To make a statement that is not true in general but only in certain special cases and not point out that it's only true in those special cases is not a "nuance," in my opinion. -- Randy Yates % "She has an IQ of 1001, she has a jumpsuit Digital Signal Labs % on, and she's also a telephone." mailto://yates@ieee.org % http://www.digitalsignallabs.com % 'Yours Truly, 2095', *Time*, ELO
Jitter generation compare to dithering
Started by ●November 5, 2009
Reply by ●November 8, 20092009-11-08
Reply by ●November 8, 20092009-11-08
Eric Jacobsen wrote:> On 11/8/2009 9:07 AM, wazerface@gmail.com wrote: >> On Nov 7, 6:02 pm, Eric Jacobsen<eric.jacob...@ieee.org> wrote: >>> On 11/7/2009 12:56 PM, Greg Heath wrote: >>> >>>> On Nov 5, 1:04 am, "BCLIM"<boonchun_...@yahoo.com> wrote: >>>>> Hi, >>>>> I have a doubt on jitter generation. Wonder is the method to >>>>> generate >>>>> jitter signal is that same as dithering generation method? >>>>> Understand that >>>>> both are usign random signal generation with different >>>>> distribution. For >>>>> example, triangular, rectangular etc. >>>> I always thought that for a signal defined over a finite length of >>>> time, >>>> jittering added a randomness to the amplitude whereas dithering >>>> added a randomness to the starting time. >>>> Hope this helps. >>>> Greg >>> I think it'd help for the OP to clarify what he means by his >>> terminology. In my experience jitter usually means random fluctuation >>> in period, (usually sampling period), while dithering usually means >>> adding small random values to an input to reduce quantization noise. >>> >>> It's also not clear whether the OP is asking about unintentional or >>> intentional jitter. Some clocking systems add jitter in order to >>> reduce spurious emissions related to the clock frequency. Is that the >>> topic of the question? >>> >>> -- >>> Eric Jacobsen >>> Minister of Algorithms >>> Abineau Communicationshttp://www.abineau.com >> >> My understanding is that dithering doesn't reduce quantization noise >> (power) in general, but de-correlates it. > > Yes, and the decorrelation allows an increase of SNR with appropriate > filtering, so it can be used to effectively reduce quantization noise. > But you're right, dithering by itself doesn't reduce quantization noise.But it can increase resolution if one has the luxury of averaging measurements. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●November 9, 20092009-11-09
>Eric Jacobsen <eric.jacobsen@ieee.org> writes: >> [...] >> Yes, and the decorrelation allows an increase of SNR with appropriate >> filtering, > >What filtering is that? If oversampling is involved, there can be >filtering. And perhaps there may be a "preference" in frequencies (e.g., >the emphasis of the human ear in frequencies near 4 kHz), which would >permit some type of filter. But in general, there is no filtering that >can be done and dither actually increases the total quantization-related >noise power. That's not to say dithering isn't worth doing, however, >because the decorrelating is a distinct advantage.There are many signal processing tasks where only the longish term result really matters. In those cases the average effect across many samples is usually all you need to see a huge win from dithering. Steve
Reply by ●November 9, 20092009-11-09
Hi Eric,
Sorry for that it's not clear. I'm interested to know the method
to generate the jitter and later inject it into a system in order to check
the jitter amount change due to the system. Basically if I have the syste
clock and would like to generate a jitter signal relative to the system
clock and inject this in to another system.
Thanks.
BC
>On 11/7/2009 12:56 PM, Greg Heath wrote:
>> On Nov 5, 1:04 am, "BCLIM"<boonchun_...@yahoo.com> wrote:
>>> Hi,
>>> I have a doubt on jitter generation. Wonder is the method to
generate
>>> jitter signal is that same as dithering generation method? Understand
that
>>> both are usign random signal generation with different distribution.
For
>>> example, triangular, rectangular etc.
>>
>> I always thought that for a signal defined over a finite length of
>> time,
>> jittering added a randomness to the amplitude whereas dithering
>> added a randomness to the starting time.
>>
>> Hope this helps.
>>
>> Greg
>
>I think it'd help for the OP to clarify what he means by his
>terminology. In my experience jitter usually means random fluctuation
>in period, (usually sampling period), while dithering usually means
>adding small random values to an input to reduce quantization noise.
>
>It's also not clear whether the OP is asking about unintentional or
>intentional jitter. Some clocking systems add jitter in order to
>reduce spurious emissions related to the clock frequency. Is that the
>topic of the question?
>
>--
>Eric Jacobsen
>Minister of Algorithms
>Abineau Communications
>http://www.abineau.com
>
Reply by ●November 9, 20092009-11-09
"steveu" <steveu@coppice.org> writes:>>Eric Jacobsen <eric.jacobsen@ieee.org> writes: >>> [...] >>> Yes, and the decorrelation allows an increase of SNR with appropriate >>> filtering, >> >>What filtering is that? If oversampling is involved, there can be >>filtering. And perhaps there may be a "preference" in frequencies (e.g., >>the emphasis of the human ear in frequencies near 4 kHz), which would >>permit some type of filter. But in general, there is no filtering that >>can be done and dither actually increases the total quantization-related >>noise power. That's not to say dithering isn't worth doing, however, >>because the decorrelating is a distinct advantage. > > There are many signal processing tasks where only the longish term result > really matters. In those cases the average effect across many samples is > usually all you need to see a huge win from dithering.Hi Steve, I am aware of this. I probably didn't state my point to Eric clearly. My point wasn't that such tasks don't exist, but that, in general (e.g., without assuming such a task), the SNR of a signal that has been dithered is actually MORE than the SNR of a signal that has not been dithered. That's not to say that the dither was bad - the decorrelation, or more intuitively, the removal of a bunch of "spikes" in the frequency spectrum (and the corresponding increase in SFDR), is usually well-worth the slight increase in overall SNR. -- Randy Yates % "...the answer lies within your soul Digital Signal Labs % 'cause no one knows which side mailto://yates@ieee.org % the coin will fall." http://www.digitalsignallabs.com % 'Big Wheels', *Out of the Blue*, ELO
Reply by ●November 9, 20092009-11-09
Randy Yates <yates@ieee.org> writes:> [...] > My point wasn't that such tasks don't exist, but that, in general (e.g., > without assuming such a task), the SNR of a signal that has been > dithered is actually MORE than the SNR of a signal that has not been > dithered.Bah! Correction: ...the SNR of a signal that has been dithered is actually LESS than the SNR of a signal that has not been dithered. and for> There are many signal processing tasks where only the longish term > result really matters. In those cases the average effect across many > samples is usually all you need to see a huge win from dithering.the correction: ...is usually well-worth the slight DECREASE in overall SNR. --Randy (Obviously I was thinking of the noise power, not the SNR...) -- Randy Yates % "The dreamer, the unwoken fool - Digital Signal Labs % in dreams, no pain will kiss the brow..." mailto://yates@ieee.org % http://www.digitalsignallabs.com % 'Eldorado Overture', *Eldorado*, ELO
Reply by ●November 9, 20092009-11-09
<begin red-faced demeanor> Randy Yates <yates@ieee.org> writes:> [...]This is misquoted:> and for > >> There are many signal processing tasks where only the longish term >> result really matters. In those cases the average effect across many >> samples is usually all you need to see a huge win from dithering. > > the correction: > > ...is usually well-worth the slight DECREASE in overall SNR.The correction is this: and for That's not to say that the dither was bad - the decorrelation, or more intuitively, the removal of a bunch of "spikes" in the frequency spectrum (and the corresponding increase in SFDR), is usually well-worth the slight increase in overall SNR. the correction: ...is usually well-worth the slight DECREASE in overall SNR. <end red-faced demeanor> <begin search for more coffee...> -- Randy Yates % "The dreamer, the unwoken fool - Digital Signal Labs % in dreams, no pain will kiss the brow..." mailto://yates@ieee.org % http://www.digitalsignallabs.com % 'Eldorado Overture', *Eldorado*, ELO
Reply by ●November 9, 20092009-11-09
On Nov 8, 3:23�am, Greg Heath <he...@alumni.brown.edu> wrote:> On Nov 7, 5:16�pm, Jerry Avins <j...@ieee.org> wrote: > > > > > > > Greg Heath wrote: > > > On Nov 5, 1:04 am, "BCLIM" <boonchun_...@yahoo.com> wrote: > > >> Hi, > > >> � �I have a doubt on jitter generation. Wonder is the method to generate > > >> jitter signal is that same as dithering generation method? Understand that > > >> both are usign random signal generation with different distribution. For > > >> example, triangular, rectangular etc. > > > > I always thought that for a signal defined over a finite length of > > > time, > > > jittering added a randomness to the amplitude whereas dithering > > > added a randomness to the starting time. > > > > Hope this helps. > > > It probably hurts. Jitter randomizes the timing of pulse edges. When you > > look at such a signal on an oscilloscope, the trace jitters. > > �Given your additional comment. it helps emphasize that > terminology is application dependent. > > In the fields of classification and regression, the terminology > is as I have indicated.Unfortunate circumstance. From the comp.ai.neural-neta FAQ: ftp://ftp.sas.com/pub/neural/FAQ3.html#A_jitter Hope this helps. Greg
Reply by ●November 9, 20092009-11-09
Wow good dithering discussion! Yes, using a pure energy metric | |^2, dithering will always increase noise power and decrease overall SNR. But whitening often eases detection / estimation / human audio perception. The real discussion is, what is the best way to choose quantization levels for representing a given signal? P. T. Boufounos and A. V. Oppenheim, "Quantization noise shaping on arbitrary frame expansion," investigates this in general. -Martin
Reply by ●November 9, 20092009-11-09
On Nov 9, 5:33�am, Greg Heath <he...@alumni.brown.edu> wrote:> On Nov 8, 3:23�am, Greg Heath <he...@alumni.brown.edu> wrote: > > > > > On Nov 7, 5:16�pm, Jerry Avins <j...@ieee.org> wrote: > > > > Greg Heath wrote: > > > > On Nov 5, 1:04 am, "BCLIM" <boonchun_...@yahoo.com> wrote: > > > >> Hi, > > > >> � �I have a doubt on jitter generation. Wonder is the method to generate > > > >> jitter signal is that same as dithering generation method? Understand that > > > >> both are usign random signal generation with different distribution. For > > > >> example, triangular, rectangular etc. > > > > > I always thought that for a signal defined over a finite length of > > > > time, > > > > jittering added a randomness to the amplitude whereas dithering > > > > added a randomness to the starting time. > > > > > Hope this helps. > > > > It probably hurts. Jitter randomizes the timing of pulse edges. When you > > > look at such a signal on an oscilloscope, the trace jitters. > > > �Given your additional comment. it helps emphasize that > > terminology is application dependent. > > > In the fields of classification and regression, the terminology > > is as I have indicated. > > Unfortunate circumstance. From the comp.ai.neural-neta FAQ: > > �ftp://ftp.sas.com/pub/neural/FAQ3.html#A_jitter > > Hope this helps. > > GregSo there exist neural net folks who made sloppy use of signal processing terminology too. Is that supposed to be a surprise to anyone? Dale B. Dalrymple
