Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Pages: [1] 2

Author Topic: Bits per bin / tones - units  (Read 5131 times)

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Bits per bin / tones - units
« on: May 06, 2018, 01:30:09 AM »

I have extracted the following info from one of my three DLink DSL-320B-Z1 modems,  line cwcc@a.4 for the first time ever using telnet, thanks to kitizen G.Dmt’s guidance.

tc> wan adsl l f

relative capacity occupation: 100%
noise margin upstream: 5.8 db
output power downstream: 17.8 dbm
attenuation upstream: 42.5 db
carrier load: number of bits per symbol(tone)
tone   0- 31: 00 00 00 08 89 99 88 88 88 87 76 65 56 55 44 44
tone  32- 63: 07 78 9a aa ca cc bc ac 8c cc cc cb bc aa cc cc
tone  64- 95: 7b bb cb b0 bb ba ba b9 a9 8a 9a 9a 97 88 99 88
tone  96-127: 78 88 86 77 67 77 45 54 55 64 55 55 45 03 53 32
tone 128-159: 34 40 00 00 00 00 00 00 00 00 00 00 00 00 00 00
tone 160-191: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
tone 192-223: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
tone 224-255: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00


Could someone help me interpreting this lot? What are the units?

I am thinking that this is actually SNR, otherwise why are there not merely 15 possible values or less?

If that is SNR, where do I find bits-per-bin, I have been digging around amongst commands in telnet with no luck. It’s a real maze.
« Last Edit: May 06, 2018, 01:53:46 AM by Weaver »
Logged

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Bits per bin / tones - units
« Reply #1 on: May 06, 2018, 01:59:34 AM »

:hmm:  Hmm . . .

Clearly hexadecimal values are being reported. Ah, yes. I believe I see it!

Sub-carrier (Tone)    Number of Bits Loaded

      00                        0
      01                        0
      02                        0
      03                        0
      04                        0
      05                        0
      06                        0
      07                        8
      08                        8
      09                        9
      10                        9
      11                        9
      12                        8
      13                        8
      14                        8
      15                        8


Etc.  ;)
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - units
« Reply #2 on: May 06, 2018, 02:13:13 AM »

Go how stupid of me, it's one digit per tone not two, years of looking at hex bytes! Thank you so much, glad someone still has a brain that works.
« Last Edit: May 06, 2018, 02:26:42 AM by Weaver »
Logged

re0

  • Reg Member
  • ***
  • Posts: 840
Re: Bits per bin / tones - units
« Reply #3 on: May 06, 2018, 02:37:05 AM »

I wanted to go a bit more in depth but I don't think my brain quite has all the knowledge required to work this out correctly. :no:

Anyway, I'll have a crack at it.

What I know is that each tone can load approx. 15 bits, which is then subject to Quadrature Amplitude Modulation (QAM). You take the potential 15 bits and times it by the symbol rate of 4000 (per second) and you end up with a maximal data rate of 60,000 bps (or 60 kbps) per tone.

I tried to calculate your approx. sync rates with the data you provided, but came up with 3252/636 Kbps. I must admit, I did just write the values manually in a spreadsheet so maybe there are some minor inaccuracies. But perhaps this doesn't take into account other factors such as interleaving?

Any thoughts?
Logged
ISP: Gigaclear - Hyperfast 900 (up to 940 Mbps symmetrical)

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - units
« Reply #4 on: May 06, 2018, 03:05:01 AM »

Have to take out FEC overheads from that number, so I believe, I only know the downstream sync rate of approx 2810kbps which is the throughput figure reduced because of those overheads.

Interleave itself doesn't slow things down surely, meaning reduces throughout, it just adds latency, but it might be that interleave happens to always be accompanied by overhead of some sort in practice.

As for overhead, I haven't yet found all the info on the values of G.992.3 framing parameters as listed in section 7.6. A lot of modems give a nice concise list of these single-character variable names B, M, T, R, D, L etc in a prominent place in the ui. No such luck with this device.

I notice a notch at tone 122. (Would that be 122 * 4.3125 kHz = 526.125 kHz, if tone #0 = 0 Hz ?) Could that be radio?

And there is another huge notch at bin 71, 306kHz. Is there anywhere you can look these frequencies up?
« Last Edit: May 06, 2018, 03:12:03 PM by burakkucat »
Logged

re0

  • Reg Member
  • ***
  • Posts: 840
Re: Bits per bin / tones - units
« Reply #5 on: May 06, 2018, 06:18:24 AM »

The FEC overheads probably accounts for the redundant [for normal use] bits. I am not sure if the depth (D, as shown on most modems) has an impact. But R (for the redunancy in Reed Solomon code) should have a slight overhead with parity bits.

On ADSL (or at least BTW lines), any sort of interleaving always comes with a delay (minimum of 8ms). The delay being higher just increses the chance of data recovery.

There's probably a command for retrieving that information from your modem. I would like to search for it but doing almost anything on this laptop is awful. So right now, I don't want to get a headache from being frustrated by it.

Tone 122 would be 526.125 kHz, as tone 0 = 0 Hz (though, this is not utilised for data). The notch could be many things. Not sure what kind of radio would be operating on this frequency (and I do not want to go looking because of what I mentioned above in regards to this laptop). But http://www.frequencyfinder.org.uk/ may provide some information. For the frequencies you've mentioned, you're going to want to look at AM.

I kinda miss having an ADSL line, even though it was awful. I wish I had taken more time when I had it to understand it much better. I am too scared to experiment with a VDSL line since there's a long way down in terms of sync speed!
Logged
ISP: Gigaclear - Hyperfast 900 (up to 940 Mbps symmetrical)

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - unit
« Reply #6 on: May 06, 2018, 06:51:15 AM »

Thank you very very much for that link.

Does anyone else have experience with pet frequencies, culprits for rf interference?
Logged

ejs

  • Kitizen
  • ****
  • Posts: 2078
Re: Bits per bin / tones - units
« Reply #7 on: May 06, 2018, 07:08:02 AM »

Bin 71 is probably the pilot tone. It can carry 2 bits on ADSL2/2+, or zero.
Logged

parkdale

  • Reg Member
  • ***
  • Posts: 597
« Last Edit: May 06, 2018, 11:31:35 AM by parkdale »
Logged
Vodafone FTTC ECI cab 40/10Mb connection / Fritz!box7590

burakkucat

  • Respected
  • Senior Kitizen
  • *
  • Posts: 38300
  • Over the Rainbow Bridge
    • The ELRepo Project
Re: Bits per bin / tones - units
« Reply #9 on: May 06, 2018, 03:26:29 PM »

I notice a notch at tone 122. (Would that be 122 * 4.3125 kHz = 526.125 kHz, if tone #0 = 0 Hz ?) Could that be radio?

And there is another huge notch at bin 71, 306kHz. Is there anywhere you can look these frequencies up?

Along with the link that you have already been given, I also make use of HF Bands - RSGB, Medium Wave Radio and Short Wave Info.

The xDSL sub-carrier width is 4.3125 kHz and, as with every xDSL variation, sub-carrier zero is half of that width, 2.15625 kHz. Hence the centre frequency of any sub-carrier is its numerical index multiplied by the width.

The BBC Radio 4 transmitters (Droitwich, Burghead and Westerglen -- all synchronised and phase-locked) radiate on 198 kHz. As a result, for virtually every xDSL circuit in the U.K., sub-carriers 45 & 46 will be affected.
Logged
:cat:  100% Linux and, previously, Unix. Co-founder of the ELRepo Project.

Please consider making a donation to support the running of this site.

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - units
« Reply #10 on: May 07, 2018, 01:04:09 AM »

* I can't seem to find anything on those frequencies, which are quite low. Any vague suggestions?

* According to wikipedia medium wave starts at 526.5 kHz and I am getting a hit at 526.125kHz (and I think posters confirmed the arithmetic was correct, B*cat, Re0 ?) so what is immediately below this edge?

*Also, I wanted to ask, how many marks out of ten does this line get?
« Last Edit: May 07, 2018, 01:56:11 AM by Weaver »
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - units
« Reply #11 on: May 07, 2018, 02:18:56 AM »

I am not having much luck digging around for 526.125kHz. NavTex is listed at 518kHz that's all that I could find. Isn’t that too definitely too far away though, unless there is more than one such allocation of something but I see no mention of such.

On Ofcom’s website I did find “Maritime mobile” and “Aeronautical radio-navigation” bands listed at 505 - 526.5kHz though.
« Last Edit: May 07, 2018, 02:28:16 AM by Weaver »
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - units
« Reply #12 on: May 07, 2018, 03:56:02 AM »

I notice that I do seem to have got away with it as far as exposure to BBC Radio 4 is concerned, perhaps Burghead and Falkirk are just way to far away - 155 mi and 184 mi respectively - and we have the serious mainland mountains in the way, although diffraction might presumably prevent any shadows.

Anyway maybe bin 71 is a pilot tone and bin 122 then is something to do with maritime comma of some sort. We do have maritime action at the Caol Loch Aillse base (anglicised as ‘Kyle’) what with submarines and things and stuff on the sea shore at Applecross (A’ Chomraich Abrach) not far from Lon Bàn.
« Last Edit: May 07, 2018, 04:17:47 AM by Weaver »
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - units
« Reply #13 on: May 07, 2018, 04:20:37 AM »

Modem #2 on line line cwcc@a.3 shows a very different story on the upstream side

output power downstream: 18.0 dbm
attenuation upstream: 42.3 db
carrier load: number of bits per symbol(tone)
tone   0- 31: 00 00 00 08 89 99 88 87 77 66 66 55 55 54 44 44
tone  32- 63: 04 57 89 99 bb ab ba 3c cc cc bb cc cc 87 cc cc
tone  64- 95: 8c ca cc b0 bb bb bb a9 aa aa 68 87 76 77 88 88
tone  96-127: 88 78 88 75 44 67 46 57 55 66 46 55 54 30 44 44
tone 128-159: 44 44 40 00 00 00 00 00 00 00 00 00 00 00 00 00
tone 160-191: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
tone 192-223: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
tone 224-255: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Logged

Weaver

  • Senior Kitizen
  • ******
  • Posts: 11459
  • Retd s/w dev; A&A; 4x7km ADSL2 lines; Firebrick
Re: Bits per bin / tones - units
« Reply #14 on: May 07, 2018, 06:57:36 AM »

I notice that bin 45 is pretty much stuffed on this line but not on the other one. The copper pair described second here, modem #2 line cwcc@a.3, is the second-activated pair in the first cable. Modem #3, line cwcc@a.4, described first, is in a second drop cable.
Logged
Pages: [1] 2