Signal attenuation vs line attenuation

I have ADSL2+ service. I'm not complain about my Internet connection however I'm very interested in details of ADSL2+ technology. Could someone explain my issues?

I've read about difference between signal attenuation and line attenuation:
http://whirlpool.net.au/wiki/ADSL_Theory_Attenuation
On this page I've read that line attenuation is measured on DSLAM. However my router give me both signal attenuation value as well as line attenuation.

Text
1 23 45 67 8	adsl: ADSL driver and PHY status Status: ShowtimeRetrain Reason: 8000Channel: FAST, Upstream rate = 1229 Kbps, Downstream rate = 15383 Kbps                 Down            UpSNRM(dB):       5.3             6.9 LATN(dB):       25.0            14.8SATN(dB):       25.0            14.1 TxPwr(dBm):     18.6            12.1

I'm curious whether SATN is depended on power output or other constraints. I've noticed my SATN for download is always the same for download as LATN, however my SATN for upload is often lower than LATN. I'm noticed also that on upstream I often have higher SNR margin and lower SATN despire higher synchronization rate, lower power output value and the same used tones. I'm very suprised by that. I suspect my router sometimes erronously report Power Output for upstream and SATN is lower if power output is higher. Do I think rightly?

In order to be accurate I'm going to give some datas. I have for example: at 1229 kbps continously SNR margin 6.8-7.0 dB and power output 12.1 dBm and SATN 14.1 dB or 1221 kbps, continously SNR margin 6.4-6.6 dB, power output 12.3 dBm and SATN 14.7 dB. Sometimes at the same power output at higher synchronization rate I have lower SATN and higher SNR margin.

Edited by konrado5 (Thu 07-Nov-13 21:00:15)

One is calculated on the attenuation at a set frequency and the other is arrived at from analysis of attenuation across all the bins for the signal.

ADSL2+ and ADSL and ADSL2 is split into discrete frequency bins, and each have their own attenuation/noise margin and bit loading. The router display pages are just a short hand for something that would give you hundreds of figures - i.e. a bin plot

Power output reports from routers are often unreliable, or affected by speed caps and power backoff procedures used to reduce cross talk.

Your upstream may perform a little oddly as you are very close to the maximum possible with ADSL2+ unless you sacrifice some downstream speed and use Annex M.

Downstream there might be a little more speed to be had, e.g. using better cabling in the home, or trying different firmware versions or different routers.

I think both line attenuation as well signal attenuation is average across all used tones because I've had 1 dB higher line attenuation when I had 4 mbps profile with very low power output (negative dBm).

I have Annex A.

Is there anything wrong with just posting in the SAME THREAD? Is that difficult?

In reply to a post by konrado5:

I think both line attenuation as well signal attenuation is average across all used tones

no, the signal attenuation is across all tones in use the line attenuation is across the whole available spectrum.

This is why on ADSL2+ the line attenuation is 3 dB high where the extra frequencies can't be used due to line length - half of the signal is lost (3 dB) in the 1-2 MHz part. It doesn't mean ADSL2+ would be slower on a long line, as the extra loss is signal you weren't going to get to use anyway.

In reply to a post by yarwell:

In reply to a post by konrado5:

I think both line attenuation as well signal attenuation is average across all used tones

no, the signal attenuation is across all tones in use the line attenuation is across the whole available spectrum.

I agree, theoretically it should be case. However it seems routers measure line attenuation as attenuation across all used tones because I've noticed at different power output profiles I've had different line attenuation measurement.

Isn't one measured as the logarithm of the quantity and the other as the logarithm of the quantity squared?