Print

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

[snadge]

Can anyone tell me, is Hlog and Attenuation the same thing regarding the Hlog stats pulled from the router? - just mrmt32 on SkyUser has (I think) labelled the Hlog graph as 'Attenuation' in the new 2504N firmware, but the plots are Negative values such as -21db etc, but Attenuation is usually shown in Positive values such as 21db... also, in DMT Tool the 'Attenuation' graph at bottom looked much like the Hlog chart but flipped upside down and the values in DMT Tool were not negative?

can anyone shed any light on this?

is the Attenuation graph in DMT Tool worked out using Hlog & SNR or something? or is Attenuation = Hlog..?

this is a screenshot from the Sky Router Tool app which he has managed to build into the routers firmware so you can see advanced stats directly in the router... as you can see, what looks to be the Hlog graph is labelled Attenuation, but in DMT Tool the same thing is flipped upside down and given positive values (see below)

[burakkucat]

I think there is some confusion present, Snadge.

The best reference I can currently lay my paws upon is the Measuring Line Characteristics on the Huawei blog posting.

[Bald_Eagle1]

My understanding is that Hlog = Attenuation over frequency.

Hlog is usually reported as negative values thus -70dB is more attenuated than -20dB.

It is often stated in terms such as "attenuated by 70dB (positive) is more highly attenuated than attenuated by 20dB (positive)

As frequency increases, attenuation increases (becomes more negative).

In the blog posting, the red areas of the graphs are the actual data, so it can be seen that Hlog initially "looks" upside down.

BTW, unfortunately I cannot see the screenshot from the Sky Router Tool app that you linked to.

[roseway]

Here's a fairly authoritative definition of Hlog:

1) Channel Characteristics Function Hlog(f) per sub-channel
The channel characteristics function H(f) is the frequency
response of the channel, i.e., amplitude magnification and
phase shift at each frequency point, which can be used for
analyzing the physical copper loop condition, for example,
determining line quality and presence of bridge taps. Its
magnitude values are depicted in a logarithmic scale, Channel
characteristics, Hlog(f).This function yields valuable
information about the physical condition of the copper loop
and its topology.

http://www.ncc.org.in/download.php?f=NCC2008/2008_P2_1.pdf

[Ezzer]

attenuation or insertion Loss is a measure of how much weaker the signal recived is compaired to the inital strength as sent. So although it's often quoted without a minus the true notation should be as a negative.

I suppose the grammer of the term loss negates the need for the minus as wouldn;t that be (wait for it) a double negative ? 

[JGO]

Don't agree Ezzer.
Surely it only needs a minus sign if you call it "gain" ?

When you talk about privatives it gets confusing, do you draw the curtains to shut out the cold or keep the heat in ?

 

 

[Bald_Eagle1]

do you draw the curtains to shut out the cold or keep the heat in ?
 

Or is the glass still half full (positive) or already half empty (negative)?

It's no use trying to be a pessimist, it wouldn't work anyway 

[Ezzer]

Your right JGO, I've got that one wrong. Should only be a minus in reference to a gain. I've contradicted myself and caused my self to come out with a double negative. 

Other wise for me its Keep the Heat in (heat energy) Or not allow the cold air to displace the hot. and dilute the heat energy.

For me heat is energy, cold isn't an enterty, more of a vacum 

[snadge]

thanks for the many replies

according to Roseway's description the two are totally different from each other... but it looks like mrmt32 on skyuser has taken the Hlog graph and named it "Attenuation" as you can see in the screenshots - yet this would be incorrect yes/no? - he is quite a clever chappy so Iam confused hehe - also the figures and profile/contours of the graph appear to be the same/similar to those of an attenuation graph in DMT (just flipped upside down)

[burakkucat]

To use the word Attenuation in place of Hlog is quite wrong. 

[snadge]

sorry if you couldnt see

here they are: (that defo looks like Hlog to me?)

[Bald_Eagle1]

According to a JDSU written document entitled JDSU-loopanalysis_an_tfs_tm_ae.pdf:-
HLog data reported during the modem initialization phase shows attenuation over frequency.

Please see the attached one page extract from that document.

[JGO]

The plots show what I would expect for an ADSL type line;  line attenuation increasing slowly with  frequency at the HF end, due to skin effect, and also loss at the LF end due to the "micro filter". The SNR reflects this (plotted the other way up !) and also shows some narrow holes which may be due to the modem notching out local broadcast stations.Looks OK to me.

[snadge]

mmm so now it is? lol

well at least thats cleared that up then... mrmt32 was right too label the Hlog graph as Attenuation

[burakkucat]

Just as correctly as saying that 'black' is 'not white'! 

(But 'not white' is not 'black', for it could be 'red'.    )