BT came out today and fixed the line, back up to speed and got some stats from the JDSU

he said someone has been fiddling in the cab and messed my line up.

ongoing stats:-

[IMG]http://i1119.photobucket.com/albums/k631/44_bh/OR_03...[/IMG]

[IMG]http://i1119.photobucket.com/albums/k631/44_bh/line_...[/IMG]

# /bin/xdslcmd info --pbParams
/bin/xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 0
Max: Upstream rate = 17598 Kbps, Downstream rate = 47380 Kbps
Path: 0, Upstream rate = 9999 Kbps, Downstream rate = 39999 Kbps

Discovery Phase (Initial) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)

Medley Phase (Final) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)

VDSL Port Details Upstream Downstream

Attainable Net Data Rate: 17598 kbps 47380 kbps
Actual Aggregate Tx Power: - 18.1 dBm 5.4 dBm

=========================================================================

VDSL Band Status U0 U1 U2 U3 D1 D2 D3
Line Attenuation(dB): 3.4 14.1 N/A N/A 7.8 17.7 N/A
Signal Attenuation(dB): 0.0 13.3 N/A N/A 7.8 17.7 N/A
SNR Margin(dB): 17.1 17.1 N/A N/A 11.0 11.0 N/A
TX Power(dBm): -29.1 -18.4 N/A N/A 1.8 2.9 N/A

#

Sync 39999D 9995U
Attenuation: 9.6 SNR: 8.9
Line Length 300meters

The origin of your sync speed and attenuation values are obvious but would you clarify that of the line length, please?

Were you told that value by an OR engineer? Did he measure it electronically with his JDSU or read it from the hardware infrastructure database, via his laptop? Or is it your own estimation?

A reliable set of those three metrics will be of use by others whose connections are behaving in a sub-optimal way.

I asked what length my D-side was he did a test on his jdsu and told me it was 300 meters, I seen the figure my self
the attenuation of 9.6 was the figure on his jdsu
The road where I think my line takes its just over 300 meters too.

these are my stats a few weeks back I would need to plug my laptop in for fresh stats

Max: Upstream rate = 17307 Kbps, Downstream rate = 48560 Kbps
Path: 0, Upstream rate = 9995 Kbps, Downstream rate = 39999 Kbps

Discovery Phase (Initial) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)
Medley Phase (Final) Band Plan
US: (0,95) (696,1183)
DS: (32,687) (1192,1627)
VDSL Port Details Upstream Downstream
Attainable Net Data Rate: 17307 kbps 48560 kbps
Actual Aggregate Tx Power: - 17.7 dBm 7.7 dBm
================================================== ==========================
VDSL Band Status U0 U1 U2 U3 D1 D2 D3
Line Attenuation(dB): 3.4 13.4 N/A N/A 7.8 16.6 N/A

Signal Attenuation(dB): 0.0 13.5 N/A N/A 7.8 16.6 N/A

SNR Margin(dB): 16.7 16.7 N/A N/A 8.5 8.5 N/A

TX Power(dBm): -28.3 -18.0 N/A N/A 6.1 2.6 N/A

Edited by lockyatlrg (Mon 03-Oct-11 14:05:54)

Thank you for clarifying the origin of the line length.

Perhaps you might like to add to your signature block something like: "Data from OR engineer's JDSU." ?

Hi Lockyatlrg..

Glad you got it sorted. and interesting to hear about your loop length.

The JDSU and similar devices can apparently estimate length using two different methods..

1) using time domain reflectometry (TDR). TDR is known in other fields such as RADAR as pulse echo sounding. It involves the injection of a shaped pulse into the end of a loop which is terminated at the other end. The time delay before the pulse echo returns to the JDSU is measured, halved, and multiplied by the velocity factor of the cable. From this calculation, loop length can be calculated very accurately. For lengths up to 1000 metres, an astonishing accuracy of +/- 0.4% can be achieved (according to manufacturer HST).

2) by measuring magnitude responses in the frequency domain, and correlating them with empirical data for the (likely) physical characteristics of the cable(s) between your NTE and the DSLAM.

From what I understand, method 1) requires a trip to the cabinet or exchange to install an appropriate termination on the loop pair before the TDR test can be performed. This termination is required to reflect the signal back so that it can be measured at the JDSU..

Probably the less accurate method 2) is used whenever possible, to reduce wear on the shoe soles.

Also, bhawkes kindly offered to test a VDSL port of DMT. I would like to take you up on that, please. Do you run Linux, as I haven't yet looked at porting the tool back to Windows, but do plan to do so, unless someone else would like to do that....

Also it would be nice to develop a more accurate algorithm for loop length estimation using the plentiful line characteristics data from the Huawei HG612. Loop length estimation is a feature in DMT, so an updated algorithm using method 2) should just slot in.

Edited by deleted (Mon 03-Oct-11 14:37:41)

Yes already done.

He did not visit the cab to perform the loop length test.

So what is the +/- tolerance on the JDSU standalone test

There is 2 ways for my line to get to my house one is 350meters by road and other is about 300 meters ish.

My cab is getting very highly populated now with VDSL connections, which is why my SNR has slipped from 12db down to 8.9 and attainable rate has dropped.

I run both Win7 (64 bit) and Ubuntu 11.04. Let me know what I can do to help.

In reply to a post by bhawkes:

I run both Win7 (64 bit) and Ubuntu 11.04. Let me know what I can do to help.

Hi BHawkes,

Thank you! That would be great.. Glad to hear that you're a Ubuntu-er as well
I will email you a tarball when it's in a fit state.

If anyone can think of any new features to add, please say.. The DMT-UX open source codebase [1] is old now, so probably the more recent (closed source) DMT for Windows has been updated a lot since then.[2]

cheers, a

[1] http://www.spida.net/projects/software/dmt-ux/index....
[2] http://translate.google.com/translate?u=http://dmt.m...

No rush! I'm away for a few days (Tue-Sat).