Print

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

[Ronski]

Here's my works connection,on an ECI cab, with BTs auto updated firmware. I'll do my home connection tonight.

Code: [Select]

# xdslcmd info --pbParams
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 1
Last initialization procedure status:   0
Max:    Upstream rate = 6763 Kbps, Downstream rate = 42072 Kbps
Bearer: 0, Upstream rate = 6605 Kbps, Downstream rate = 36529 Kbps

Discovery Phase (Initial) Band Plan
US: (6,31) (882,1193) (1984,2770)
DS: (33,857) (1218,1959) (2795,4083)
Medley Phase (Final) Band Plan
US: (6,31) (882,1193)
DS: (41,857) (1218,1959)
                  VDSL Port Details               Upstream                Downstream
Attainable Net Data Rate:            6763 kbps              42072 kbps
Actual Aggregate Tx Power:             0.9 dBm               11.0 dBm
====================================================================================
        VDSL Band Status                U0              U1              U2     U3               U4              D1              D2              D3
  Line Attenuation(dB):  3.8     41.6    60.3     N/A     N/A    19.1    52.5   85.2
Signal Attenuation(dB):  3.8     41.6     N/A     N/A     N/A    19.9    52.1    N/A
                SNR Margin(dB):  6.6     6.3      N/A     N/A     N/A    6.5    6.5       N/A
                 TX Power(dBm): -3.5    -1.1      N/A     N/A     N/A    8.4    7.4       N/A
#

Code: [Select]

   70 -10.5000

[Ronski]

And here from my home connection on ECI cab with BT's auto updated firmware

Code: [Select]

# xdslcmd info --pbParams
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 0
Last initialization procedure status: 0
Max: Upstream rate = 10502 Kbps, Downstream rate = 46884 Kbps
Bearer: 0, Upstream rate = 8868 Kbps, Downstream rate = 46752 Kbps

Discovery Phase (Initial) Band Plan
US: (6,31) (882,1193) (1984,2770)
DS: (33,857) (1218,1959) (2795,4083)
Medley Phase (Final) Band Plan
US: (6,31) (882,1193) (1984,2712)
DS: (41,857) (1218,1959) (2795,3924)
  VDSL Port Details   Upstream   Downstream
Attainable Net Data Rate:     10502 kbps     46884 kbps
Actual Aggregate Tx Power:        6.6 dBm      10.9 dBm
====================================================================================
VDSL Band Status U0 U1 U2 U3 U4 D1 D2 D3
  Line Attenuation(dB): 2.2 32.7 50.3   N/A   N/A 16.1 42.3 66.2
Signal Attenuation(dB): 2.2 32.6 49.8   N/A   N/A 21.3 41.9 66.6
SNR Margin(dB): 7.2 7.0 7.5   N/A   N/A 6.3 6.3 6.3
TX Power(dBm): -3.5 -9.1 6.2   N/A   N/A 8.7 7.4 -4.9

Code: [Select]

   70 -8.3750

[burakkucat]

Thank you, everyone, for your contributions. I now have sufficient data to consider if my ideas are meaningful . . . (Any further data will be happily accepted and "munged" with the previous.)

A brief explanation.

I have created an ASCII textual file with five columns. The columns are --

Code: [Select]

Max Max
US DS US DS Hlog @ 70

Sorting by column one --

Code: [Select]

[Duo2 ~]$ sort -k1nr data.txt
35552 88628 20000 79987 -5.0000
32741 92264 19999 79442 -3.6875
27326 67876 20000 67405 -96.2500
27304 66276 20000 66514 -4.5000
24566 62440 10000 39998 -4.7500
18587 64856 18519 64859 -5.6875
16017 72216 14609 72216 -7.0000
10502 46884 8868 46752 -8.3750
 8126 48448 7792 39998 -10.7500
 6842 44492 6824 38339 -10.3750
 6763 42072 6605 36529 -10.5000
 5223 26948 5308 28160 -13.3750
 4983 23084 4877 20695 -12.2500
  675 17776   718 17982 -17.3750
[Duo2 ~]$

Sorting by column two --

Code: [Select]

[Duo2 ~]$ sort -k2nr data.txt
32741 92264 19999 79442 -3.6875
35552 88628 20000 79987 -5.0000
16017 72216 14609 72216 -7.0000
27326 67876 20000 67405 -96.2500
27304 66276 20000 66514 -4.5000
18587 64856 18519 64859 -5.6875
24566 62440 10000 39998 -4.7500
 8126 48448 7792 39998 -10.7500
10502 46884 8868 46752 -8.3750
 6842 44492 6824 38339 -10.3750
 6763 42072 6605 36529 -10.5000
 5223 26948 5308 28160 -13.3750
 4983 23084 4877 20695 -12.2500
  675 17776   718 17982 -17.3750
[Duo2 ~]$

Sorting by column five --

Code: [Select]

[Duo2 ~]$ sort -k5nr data.txt
32741 92264 19999 79442 -3.6875
27304 66276 20000 66514 -4.5000
24566 62440 10000 39998 -4.7500
35552 88628 20000 79987 -5.0000
18587 64856 18519 64859 -5.6875
16017 72216 14609 72216 -7.0000
10502 46884 8868 46752 -8.3750
 6842 44492 6824 38339 -10.3750
 6763 42072 6605 36529 -10.5000
 8126 48448 7792 39998 -10.7500
 4983 23084 4877 20695 -12.2500
 5223 26948 5308 28160 -13.3750
  675 17776   718 17982 -17.3750
27326 67876 20000 67405 -96.2500
[Duo2 ~]$

Hmm . . .

Why did I decide on Tone 70? Let's consider some frequencies --

Tone 69:        69 x 4.3125 kHz = 297.5625 kHz
Tone 70:        70 x 4.3125 kHz = 301.8750 kHz

There is no prize available for the first person who concludes that I am interested in 300 kHz (or there about). But why 300 kHz?

Consider the screen from the Management and Diagnostic Console of a 2Wire device, attached below. Under the General Information heading, notice the Downstream Atten. at 300 kHz: line. That is my reason. 

What am I attempting to discover / quantify? Some relationship between the maximum synchronisation speed ( US / DS ) and another parameter, obtainable from the active CPE, which can be related to the electrical length of the pair.

Perhaps I need to incorporate the power output along with the Hlog function for Tone 70 and then see if a relationship with the maximum synchronisation speed is observable.   

Another thought. We invoke an xdslcmd info --Hlog command at the busybox shell and capture the output into a file called, say, hlog.txt of 4096 lines. Rather than using the value of Tone 70, why not use the average value of all possible 4096 tones?

Using the data that I have previously harvested from my neighbour's (Gordon's) line --

Code: [Select]

[Duo2 ~]$ awk '{ X += $2 } END { print "Summed Total =", X, "\nNumber of Tones =", NR, "\nAverage Value =", X / NR }' Hlog.txt
Summed Total = -207507
Number of Tones = 4096
Average Value = -50.6609
[Duo2 ~]$

Comments / suggestions / observations, etc, are welcomed. 

[pettaw]

If you still need more data, here is a my data from Beattie latest updated firmware. I haven't flashed the latest GUI back yet.

Code: [Select]

xdslcmd info --pbParams
xdslcmd: ADSL driver and PHY status
Status: Showtime
Retrain Reason: 1
Last initialization procedure status: 0
Max: Upstream rate = 21534 Kbps, Downstream rate = 67304 Kbps
Bearer: 0, Upstream rate = 19999 Kbps, Downstream rate = 67532 Kbps

Discovery Phase (Initial) Band Plan
US: (7,32) (871,1205) (1972,2782)
DS: (33,859) (1216,1961) (2793,3970)
Medley Phase (Final) Band Plan
US: (7,32) (871,1205) (1972,2782)
DS: (33,859) (1216,1961) (2793,3970)
  VDSL Port Details   Upstream   Downstream
Attainable Net Data Rate:     21534 kbps     67304 kbps
Actual Aggregate Tx Power:        7.4 dBm      13.4 dBm
====================================================================================
VDSL Band Status U0 U1 U2 U3 U4 D1 D2 D3
  Line Attenuation(dB): 5.1 26.2 39.2   N/A   N/A 13.3 32.5 50.0
Signal Attenuation(dB): 5.1 25.2 38.3   N/A   N/A 18.4 32.2 50.0
SNR Margin(dB): 6.8 7.1 7.0   N/A   N/A 6.2 6.1 6.2
TX Power(dBm): 0.1 -17.0 6.5   N/A   N/A 10.3 7.7 7.2



Code: [Select]

70 -6.5000

[burakkucat]

Thank you. All data is happily accepted. 

[Greybeard33]

Interesting results. One observation: the "scatter" of the data points around a smooth curve might well be because the snapshots of the various lines have been taken at random times of day. Certainly on my line the max attainable rates in the evening are considerably lower than in the daytime. Likewise sync speeds vary with the time of day of the last retrain, whereas the Tone 70 attenuation appears to be much the same regardless.

For future data acquisition, might it be worth specifying a preferred time of day for the pbParams snapshot? And requesting that the time of the snapshot is also supplied?

[burakkucat]

Thank you for your thoughts, Gb33.

The question, of which I keep asking myself, is "Is this actually meaningful?" To be honest, I'm not sure. 

As for the variation of the maximum attainable rate according to time of day, that is the reason for maintaining columns 3 & 4 as well as columns 1 & 2.    I am hoping to see some sort of trend when comparing either columns 1 & 2 with column 5 or columns 3 & 4 with column 5.

I am toying with the idea of incorporating a function of the "value" at Tone 70 with the power output for that frequency and using that function as the data points for the "x-axis" of the current thought-process graph. When looking at your value, for example, what do we actually see?

70      -10.7500

Tone 70 is virtually 300 kHz -- in other words a value with the unit of s^-1.

But what is "-10.7500"? I can accept the minus sign, as we are discussing attenuation. So "10.7500" what? As it is one datum point in a stream of output that results from an "xdslcmd info --Hlog" request, it must be fair to assume the relationship y = log x and its converse 10^y = x are involved. But what actually is that entity? What is "10.7500", with a given negative sign to signify attenuation? How could that entity be "munged" with the power output (dBm)?

I'd be happy to see suggestions from anyone. 

[Greybeard33]

As I understand it, Hlog simply gives the attenuation of each tone in dB, as measured during training (log because it is a logarithmic scale). So, for my line, the attenuation was 10.75dB at 300kHz.

Subtracting the attenuation from the transmitted power of the tone should give the received signal level in dBm. The difference between this and the Quiet Line Noise (QLN) of the tone should be the Signal to Noise Ratio (SNR) in dB for that particular tone.

Of course, Hlog and QLN are only measured during training, whereas changes in SNR, e.g. due to varying levels of interference, are monitored dynamically.

[burakkucat]

Thank you for that clarification. 

I've performed some more data manipulation. The table now consists of eight columns, as follows --

Code: [Select]

Max Max Hlog
US DS US DS Power @ 70 Power + Hlog Power - Hlog

Sorting  by column seven gives --

Code: [Select]

[Duo2 ~]$ sort -k7nr data.txt
27304 66276 20000 66514 10.8 -4.5000   6.3 15.3
18587 64856 18519 64859 11.3 -5.6875   5.6125 16.9875
35552 88628 20000 79987 9.6 -5.0000   4.6 14.6
32741 92264 19999 79442 8.1 -3.6875   4.4125 11.7875
21534 67304 19999 67532 10.3 -6.5000   3.8 16.8
24566 62440 10000 39998 8.1 -4.7500   3.35 12.85
16017 72216 14609 72216 10.1 -7.0000   3.1 17.1
10502 46884 8868 46752 8.7 -8.3750   0.325 17.075
 4983 23084 4877 20695 11.7 -12.2500 -0.55 23.95
 8126 48448 7792 39998 9.4 -10.7500 -1.35 20.15
 6842 44492 6824 38339 8.3 -10.3750 -2.075 18.675
 6763 42072 6605 36529 8.4 -10.5000 -2.1 18.9
  675 17776   718 17982 11.0 -17.3750 -6.375 28.375
 5223 26948 5308 28160 6.1 -13.3750 -7.275 19.475
27326 67876 20000 67405 7.4 -96.2500 -88.85 103.65

Whilst sorting by column eight gives --

Code: [Select]

[Duo2 ~]$ sort -k8nr data.txt
27326 67876 20000 67405 7.4 -96.2500 -88.85 103.65
  675 17776   718 17982 11.0 -17.3750 -6.375 28.375
 4983 23084 4877 20695 11.7 -12.2500 -0.55 23.95
 8126 48448 7792 39998 9.4 -10.7500 -1.35 20.15
 5223 26948 5308 28160 6.1 -13.3750 -7.275 19.475
 6763 42072 6605 36529 8.4 -10.5000 -2.1 18.9
 6842 44492 6824 38339 8.3 -10.3750 -2.075 18.675
16017 72216 14609 72216 10.1 -7.0000   3.1 17.1
10502 46884 8868 46752 8.7 -8.3750   0.325 17.075
18587 64856 18519 64859 11.3 -5.6875   5.6125 16.9875
21534 67304 19999 67532 10.3 -6.5000   3.8 16.8
27304 66276 20000 66514 10.8 -4.5000   6.3 15.3
35552 88628 20000 79987 9.6 -5.0000   4.6 14.6
24566 62440 10000 39998 8.1 -4.7500   3.35 12.85
32741 92264 19999 79442 8.1 -3.6875   4.4125 11.7875

Hmm . . . Perhaps I should "sleep on it". 

[Greybeard33]

Hmm. I think one difficulty is that the "Power" column is the "Actual Aggregate Transmit Power" for all the tones in the D1 band added together, whereas the Hlog figures are just for Tone 70. I doubt that there is any simple relationship between the aggregate power and the power transmitted on each individual subcarrier, particularly since some of the D1 tones are subject to a power cutback that may vary between lines.

Unfortunately the modem cannot tell us the transmit power for individual tones. I guess it might be possible to estimate it from the QLN and SNR, although the QLN is a historic snapshot taken during retrain and the noise level may have changed since.

Asbokid's blog gives a comprehensive dissertation on the derivation of all these line parameters at http://huaweihg612hacking.wordpress.com/2011/10/01/measuring-line-characteristics-on-the-huawei/, although the complexities make my creaky old brain cells hurt! 

[burakkucat]

Asbokid's blog gives a comprehensive dissertation on the derivation of all these line parameters at http://huaweihg612hacking.wordpress.com/2011/10/01/measuring-line-characteristics-on-the-huawei/, although the complexities make my creaky old brain cells hurt! 

You and me, both. 

On more than one occasion I have attempted to gain a modicum of understanding only to awake, some time later, with my nose on the keyboard!