Jump to content

Talk:Pulse-density modulation

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

Unsorted text

[edit]

I'm not sure if the Perl code belongs in the article. I think it adds a greater understanding of the algorithm that creates pdm so I put it in their. If consensus is that it doesn't belong in the article I would be fine with moving it to the talk page so it would still be accessable to those interested. --Starx 03:22, 14 May 2004 (UTC)[reply]

The Perl code was very helpful for me.
Yet shouldn't the qe be clamped between -1 and 1?
I implemented a version in scala which supports higher orders. Perhaps it can be added.
def toPdmNew(input: IndexedSeq[Double], oversample_rate: Int, order: Int) = {
input.map(sample => (0 until oversample_rate).map(_ => sample)).flatten
.scanLeft((0 until order).map(_ => 0.0)){ (acc_prev, x_n) =>
val y = ((acc_prev.last >= 0).toInt * 2 - 1)
val e = y - acc_prev.last
acc_prev.scanLeft(x_n)((n_1, n) => (n_1 + n - y).clamp(-1, 1)).tail
}.map(x => (x.last>=0, x.last))
}
Jaspervb (talk) 22:53, 30 May 2024 (UTC)[reply]

Starx asked me to roll back the page history so that it would not contain edit summaries by a vandal which appeared to be hostile and obscene. I couldn't figure out how to do that, so I moved the page elsewhere, then deleted it, then re-created it with the same content. I'm going to research this further, but for now I think this is acceptable for this page. All of the relevant page history consisted of creation and edits by Starx, followed by several cycles of vandalism and reversion. Advice welcome. Dpbsmith 22:14, 26 May 2004 (UTC)[reply]

I've gone ahead and history merged the first few edits back into the page; they were at the title User:Dpbsmith/foo. The actual vandalism seems to have been deleted, but I can find no mention of that operation being carried out at the relevant village pump discussion. It's nowhere near as important as the first few edits of the article, though. Graham87 15:03, 4 June 2011 (UTC)[reply]

Does human hearing (or perhaps the nervous system generally?) work this way? — Preceding unsigned comment added by 222.8.82.204 (talk) 22:30, 26 May 2004 (UTC)[reply]

What "way" are you refering to? --Starx 03:53, 27 May 2004 (UTC)[reply]

Relation with other modulations

[edit]

Why cannot the PDM be considered as a 1-bit PCM? PCM is also amplitude-proportional (the higher pulses correspond to higher input). Both are lowpass-filtered for restavration. Afterall, the PDM is called a SDM by some and the feature of being proportional to the input is not inherent to DMs. --Javalenok 19:03, 16 April 2006 (UTC)[reply]

The article claims that the PDM is a ∑∆ modulation. What is the difference between them? --Javalenok 08:26, 4 June 2006 (UTC)[reply]

I see no difference between them, and I found a reference that claims that they are the same. --TedPavlic | talk 13:46, 18 February 2008 (UTC)[reply]
Yes it is identical to the output of a 1-bit DAC. I added a sentence in the introduction to this effect.86.142.20.148 (talk) 21:23, 25 August 2012 (UTC)[reply]

What about PFM modulation (with discrete f steps), I implemented modulation and can not decide which one it is. Maybe page comparing this types of modulation would help. —Preceding unsigned comment added by 217.67.27.58 (talk) 09:45, 7 October 2010 (UTC)[reply]

merge

[edit]
The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
Closed. No consensus. KeithbobTalk 20:42, 27 July 2014 (UTC)[reply]

I suggest merging delta-sigma modulation into pulse-density modulation. As far as I can tell, they are exact synonyms. --68.0.124.33 (talk) 15:18, 3 July 2008 (UTC)[reply]

I agree that the two names are exact synonyms. My concern is that someone with an interest in delta-sigma analog to digital conversion would no longer find section 1 of the DSM article using a search engine. It may be appropriate to merge the two articles if section 1 is split off as an article on delta-sigma analog to digital conversion with a reference to the pulse density article. The theoretical treatments are sufficiently different that this separation may be in the best interest of the non-specialist enquirer but the breakdown of circuit waveforms given in section 1 may also be of interest to those only interested in the modulation case since the technology is very similar. I say this because none of the treatments for the modulation case I have seen give explicit circuit diagrams with corresponding timing diagrams. This limits the accessability of the discussion to those already familiar with the jargon which somewhat misses the point of an encyclopedia article. Even so I leave that to the people most concerned with discussing modulation techniques.Puffingbilly (talk) 05:52, 5 July 2008 (UTC)[reply]


merge

1) PCM refers to pulse-code-modulation and as such is not a modulation scheme. As described above, PCM is amplitude shift keying (though they called it PCM in the early days to keep it similar to AM, FM).

2) after reading the article on sigma-delta modulation (more common terminology than delta-sigma but that's a personal preference thing), the article on DSM is inaccurate as sigma-delta modulation is not an ADC/DAC technology but is used much more widely than that (I recently made a modification to suggest this). SDM produces a PDM output. A first order PDM signal is spectrally no different that pulse-width modulation (though with a possible variable period). The sigma-delta modulation page should include this content. Most of the other page includes implementations which is fine, but that knowledge is missing from the other page.

So conclusion lots of similarity, this page would compact significantly if merged with the other page. —Preceding unsigned comment added by Rjf ie (talkcontribs) 18:22, 6 July 2008 (UTC)[reply]


merge the other way

I also agree that a merge should be made, but I think that it is the pulse-density modulation that should be merged into the delta-sigma modulation article. In the recent research literature (IEEE etc.), the terms sigma-delta or delta-sigma appear much more often than the term pulse-density modulations. Hanspi (talk) 06:34, 10 July 2008 (UTC)[reply]

I agree with Hanspi. Don't merge the more important search topics--sigma-delta and delta-sigma--into a more obscure, roll-up category. Jdonovan43 (talk) 12:26, 6 November 2008 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

multiplexing of stereo onto a single data wire

[edit]

Hi Came here trying to understand Akustica Products -

pulse density modulation (PDM), a single-bit digital output stream which enables the multiplexing of stereo microphone data onto a single data wire

How?

  • Interleaved pulses LRLRLRLRLRLRLR
  • Positive and negative pulses ? ie Tri-state bus ?

--195.137.93.171 (talk) 07:27, 21 December 2011 (UTC)[reply]

I added a few words to the article about transmitting PDM stereo audio over a single data wire. Does that answer your question? I also added links to references that go into more detail -- feel free to add more details to this article, if you think it will help our readers.

It's interleaved LRLRLRLRLRLRLR..., just as you suspected.

I worry that the phrase "single data wire" is a little misleading unless the article mentions the other wires -- the master clock wire and the return wire(s).

p.s.: In general, when you want to find out details about a chip, it is often enlightening to read that chip's datasheet. --DavidCary (talk) 04:30, 7 August 2014 (UTC)[reply]

Diagram

[edit]

The modulation diagram is great, perhaps it would be more helpful if the bitstream above it was exactly aligned?
I think it needs to be part of the actual image to avoid scaling mismatches, but that is beyond my ability. — Preceding unsigned comment added by 207.81.124.137 (talk) 02:55, 12 January 2022 (UTC)[reply]