Broadband Related > Broadband Hardware
Notes on the xdslcmd in Broadcom based modems
JustAnother:
Reserved post 4.
kitz:
Very interesting article thank you for posting
>> Annex B:
That has me puzzled because thats the standard for adsl over ISDN and used in certain countries, whereas I thought we used Annex A / Annex M in the uk.
Out of curiousity does the standard firmware connect using annex B or is it just on the unlocked firmware? My understanding is that the annex defines the mapping available for the downstream/upstream tones for use. Annex A defines which tones are available for the upstream reserving those needed for POTS. Im assuming if the router is set at Annex B, yet the standard used by BT is Annex A, the MSAN will just over-ride this.
>> OHF and OHFErr: TODO – Unknown.
Im guessing here, but Frame OverHeads? These are a checksum byte to ensure the data is being transmitted/recieved properly, working in a similar way to HEC's or CRC's.
DSL data frames (see bottom of page here ) have checkbytes, and the OHFErr would then be the amount of errors detected and the data frame had to be retransmitted.
They certainly appear to be some sort of frame counter.
>> Delay: I think this is the delay introducted by interleaving on the line.
See http://www.kitz.co.uk/adsl/interleaving.htm
--- Quote ---Interleave Delay defines the mapping (relative spacing) between subsequent input bytes at the interleave input and their placement in the bit stream at the interleave output.
--- End quote ---
>> As the interleaving level can be differed,
The maximum interleave delay is something that can be configured by the SP* on the MSAN.
(SP being the owner of the MSAN say Be/Sky etc)
btw.. not sure if youve seen this http://www.kitz.co.uk/adsl/linestats_errors.htm which shows some of the common counters such as RS and RSCorr UAS etc
Bald_Eagle1:
Hi kitz,
--- Quote from: kitz on December 19, 2011, 09:38:17 AM ---Very interesting article thank you for posting
>> Annex B:
That has me puzzled because thats the standard for adsl over ISDN and used in certain countries, whereas I thought we used Annex A / Annex M in the uk.
Out of curiousity does the standard firmware connect using annex B or is it just on the unlocked firmware? My understanding is that the annex defines the mapping available for the downstream/upstream tones for use. Annex A defines which tones are available for the upstream reserving those needed for POTS. Im assuming if the router is set at Annex B, yet the standard used by BT is Annex A, the MSAN will just over-ride this.
--- End quote ---
Here's a snippet from my unlocked modem:-
--- Code: ---# xdslcmd info --stats
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 0
Max: Upstream rate = 6210 Kbps, Downstream rate = 31968 Kbps
Path: 0, Upstream rate = 6311 Kbps, Downstream rate = 27189 Kbps
Link Power State: L0
Mode: VDSL2 Annex B
VDSL2 Profile: Profile 17a
TPS-TC: PTM Mode
Trellis: U:ON /D:ON
Line Status: No Defect
Training Status: Showtime
Down Up
SNR (dB): 3.7 5.5
Attn(dB): 0.0 0.0
Pwr(dBm): 12.3 6.2
VDSL2 framing
Path 0
B: 31 191
M: 1 1
T: 64 40
R: 12 12
S: 0.0374 0.9657
L: 9401 1690
D: 1743 1
I: 44 102
N: 44 204
Counters
Path 0
OHF: 22277373 1375498
OHFErr: 3274 217
RS: 4259640857 2267123
RSCorr: 3490669 947
RSUnCorr: 363316 0
Path 0
HEC: 70735 0
--- End code ---
Elsewhere, this is mentioned:-
G.992.3_Annex_K_PTM
Does that help at all?
Paul.
kitz:
Another Annex B
>>> G.992.3_Annex_K_PTM
Where was that??? :o Now that is VERY strange - because G.992.3 is actually adsl2!! ???
VDSL should be showing as G993.x
Basically the G number defines the type of DSL, whilst the Annex defines the spectral mask in use.
As I mentioned earlier, its the Annex which maps out which tones can be used for (DMT) bitloading and these masks set the maximum bits which can be used in each particular bin (hence why we sometimes see curved bit loading graphs particulary on the upstream and lower end of the downstream).
This is a typical example of an Annex A Mask
- Note the inverted 'U' shape for bitloading
An Annex M Mask simply shunts the lower downstream tones (those that you see with not as much bit loading) over to be available for upstream. These crossover tones dont get the full bitloading to keep downstream v upstream crosstalk between the 2 technologies to a minimum on the MSAN.
TBH I dont know much about Annex B as I didnt think it was in use in the UK (but then again I dont have any personal experience with VDSL), but off the top of my head it gives some of the POTS reserved tones in Annex A over to the upstream on Annex B
Bald_Eagle1:
Hi kitz,
--- Quote from: kitz on December 19, 2011, 04:12:09 PM ---Another Annex B
>>> G.992.3_Annex_K_PTM
Where was that??? :o Now that is VERY strange - because G.992.3 is actually adsl2!! ???
VDSL should be showing as G993.x
--- End quote ---
It comes from the modem's URL http://192.168.1.1/html/status/xdslStatus.asp
The full details are:-
--- Code: ---var DSLCfg = new Array(newstDsl"InternetGatewayDevice.WANDevice.1.WANDSLInterfaceConfig",
"Up","VDSL2","","6311","27189","0","0","6348","32044","57","38","0","0","62","123","G.992.3_Annex_K_PTM"),null);
var DSLStats = new Array(new stStats("InternetGatewayDevice.WANDevice.1.WANDSLInterfaceConfig.Stats.Showtime",
"113126","0","585276","0","6252","290","0","0","0","0","0","0"),null); var DslUpTime = "117811";
var time = 0;
--- End code ---
It is quite possible that the modem is displaying G.992.3 incorrectly, instead of G.993.2.
It does display some other data back-to-front or a bit "mixed up".
It would appear that a number of bugs may be present in the Broadcom software. The general impression from those that understand these things is that it is rather poorly written.
I see that ITU-T G993.2 does cover "Very high speed digital subscriber line transceivers 2 (VDSL2)"
&
ITU-T G.992.3 covers "Asymmetric Digital Subscriber Line (ADSL) transceivers 2 (ADSL2)"
Paul
EDIT:-
BT's variant of VDSL2 does "optionally" use upstream band U0 (tones 0 to 95).
My understanding is that that US band is used in the UK to assist somewhat with US over longer line lengths.
& some of that actually overlaps with DS band D1 (tones 32 to 859).
I have a "longer" line length, & I see quite a fair amount of bitloading at these lower frequency bands (see attached graph) The blue sections show the "shared" tones
[attachment deleted by admin]
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version