Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Oranged]

Does anybody know what could be causing permanently low bitloading at 383.8125 KHz on Tone 89 ?

It has always been like this even when I was with Orange on ADSL LLU and is the same now with O2 ADSL2+.

I never see any bitswapping nor have I ever seen anything other than 2 bits logged in Routerstats, it's like bitswapping has seized up on Tone 89.

I don't believe there are any radio stations broadcasting on/near that frequency and I can't identify any other equipment that could be on/near that frequency. To my knowledge there are no amateur radio users locally.

I live in the north Bristol area and establishments within 3Km (straight line) are MoD Abbeywood, Frenchay Hospital, University of West of England and Filton airfield. Does anybody know if equipment used in any of these could cause this ?

I would say that this problem has no effect on my connection.......I'm just very curious 

[waltergmw]

Hi Oranged,

I suspected that Filton might have something to do with it especially as any Radars, and their harmonics, are usually quite powerful radio sources.
However looking here doe not yield any accurate fit for Filton but I've no idea when this data was generated.

http://www.radios-uk.com/airfrequencies.htm

Looking here your frequency is in a military air band.

http://www.radios-uk.com/scanners/frequencies.htm

Sorry I can't spot exactly what it is.

Kind regards,
Walter

[Oranged]

Thanks for your thoughts Walter.

Filton isn't a "commercial" airfield as such. Royal Mail have a regional depot there, the Police helicopter is based there, private traffic and of course Bae and Rolls Royce use it and to my knowledge there's no sophisticated ILS etc just basic radar. Also the frequencies shown in that link are in MHz.....I guess ?

My suspicions once were the MoD Abbeywood but that military frequency is again in MHz not KHz.

So I'm at a loss......what can be operating in the medium wave band between 379.5KHz Tone 88 and 388.125KHz Tone 90 ?

The bitloading from Tone 33 to 104 is either 14 or 15 except Tone 89 and it's the same when I hook up a different make router.

[jeffbb]

Hi
You don't mention what the snr margin is for that tone if the margin is high then its simply that during the negotiations when starting up this tone was not allocated . every tone may NOT be filled to its full potential for a variety of reasons .
1  Spare capacity required for bit swapping ,
 2 not required to be filled ,remember that synch rates are in blocks of 512Kbps  so even though a tone is available it would not make up a valid synch rate .
an example is one range is from 4000 to 4512 kbps  the next valid synch rate is 4544Kbps . So if during negotiations the max rate possible was 4543Kbps then 7 or 8 bits(4Kbps per bit) would be dropped to allow a synch rate of4512 to be achieved . if say 3 of those bits were not loaded on tone 89 then the SNR margin for that tone would be  ~9db higher than the others around it

If the SNR margin is low then that is  when you would suspect some noise on that frequency.
Normally radio (MW) affects not only the one tone  but also say 2 to 3 tones on either sides forming a rough V
On the bitswap /tone you would normally see a peak on that tone ,but only if the condition was changing 
MW interference tends to vary depending on time/weather .

Some times get this with one or two tones ,they normally are back to normal after a reboot .

Regards Jeff

[GunJack]

that frequency is somewhere around airfield navigation aids, so wouldn't rule out Filton as the source....

[Oranged]

Normally radio (MW) affects not only the one tone  but also say 2 to 3 tones on either sides forming a rough V
On the bitswap /tone you would normally see a peak on that tone ,but only if the condition was changing 

Yes I get that effect on Tone 164 694.3125KHz which is Radio 5 Live on 693KHz, Tones 158 to 164 form a V.

I understand your suggestion regarding SNR but......it has always been like that and I wouldn't expect SNR to permanently influence bitloading....I'm talking more than 6 years from the time I started with Orange.

[Oranged]

that frequency is somewhere around airfield navigation aids, so wouldn't rule out Filton as the source....

Also Frenchay Hospital has a helipad and I guess that has some form of radar/beacon transmission unless they use the Filton radar/beacon which is not that far away.

[geep]

I'm quite close to Stansted and have a big hole at tone 83, being 353.625 KHz. But not always.

< http://pgg999.co.uk/stats/st546.bitloading.html

Cheers,
Peter

[lemzip]

I have noticed that most of the users on say TT at around my speed have the following tones closed

Gap1 474kHz-479kHz (Tone110)
Gap2 548kHz-552kHz (Tone127)
Gap3 811kHz-815kHz (Tone188)
Gap4 824kHz-828kHz (Tone191)
Gap5 1048kHz-1052kHz (Tone243)
Gap6 1255kHz-1259kHz (Tone291)
Gap7 1501kHz-1505kHz (Tone348)




If i sync over 17mbit, these tones get used, no idea why but we all seem to get it, maybe something set on the msan? Could yours be similar?

[renluop]

Sorry to butt in with what is somewhat irrelevant.
Jeffbb wrote

synch rates are in blocks of 512Kbps

It's likely that I am getting confused, but how comes it then that my current sync is 6319?

[jeffbb]

Hi

its in the range of 6240 to 6784 . the next valid block will start at 6816.

Regards Jeff

[renluop]

Ah so I misunderstood. Quel surprise.

[Oranged]

You don't mention what the snr margin is for that tone if the margin is high then its simply that during the negotiations when starting up this tone was not allocated.

How do I find out the SNRM at Tone 89 ?

I have O2's TG585v7 and a Belkin 7633 but, to my knowledge, neither display that data in DMT or Routerstats.

I don't see that middle DMT graph that's shown in lemzip's post above.

[jeffbb]

Hi
quote How do I find out the SNRM at Tone 89 ?

In RS IT is displayed in the BITS and tone graph .Works for netgear ,not familiar with your routers or how much info you can get

In DMT there is a display of SNR per tone ,generally the 2 graphs have a similar shape

quote I don't see that middle DMT graph that's shown in lemzip's post above.

I get it ok . Unfortunately I am no expert with DMT . I find it very cumbersome . Sometimes I don't get the 3 Graphs ,no real Idea why. Can you post your DMT results and perhaps some differences in settings can be seen?.
Regards Jeff
edit  missed end of graph with scale  0 to 20 for snr Margin



[attachment deleted by admin]

[GunJack]

jeff - in both RS and DMT, what results you get really depend on your router and even what fw you use on it. E.g. my D-Link 2640B on v4.01 fw will only work with RS if connected via telnet and it won't work at all with RSLite (despite Mr Owen trying several times to find a fix), but when I was using v3.06 fw I couldn't use RS and could only use RSLite (but now I can telnet I havent been back and tried it again on 3.06).

Just to clarify, the RS Bits/tone graph incorporates the s/n per tone in it (the line graph superimposed over the bitloading part), but on DMT the middle graph is just the s/n per tone, so it's easier to see (in snr terms) how each tone is performing.

..hope this doesn't come over, like, granny and eggs, IYSWIM