I would be interested to know why the line attenuation figures differ so much for upstream and downstream communication when the same copper/fibre cable is being used for the link to my house?

enter image description here

PS Apologies for the tag “routing” as I could not find a suitable one for my question.

  • 3
    Likely because you need the signal strength measured from the receiving end to determine how much attenuation is present. The upstream device may not be sending this information to your device, so it is displaying a value of 0. Without knowing the equipment on both sides of the connection, this is only a guess (as will be any answer). – YLearn Mar 6 '18 at 9:23
  • @YLearn So not a trivial matter? I am happy with the speeds provided by Sky but was interested in the statistics provided by the router which Sky use to check whether the speeds are “as advertised”. – Farcher Mar 6 '18 at 9:27
  • Unfortunately, questions about home networking, consumer-grade devices, and networks over which you have no direct control are all off-topic here. You could try to ask this question on Super User. – Ron Maupin Mar 6 '18 at 14:15
  • @RonMaupin As a first time user of this arm of Stack Exchange I was unaware of the rules. I am sorry to have caused you any inconvenience although I must say that I have got t very satisfactory answers and a useful comment. – Farcher Mar 6 '18 at 14:35

Attenuation is dependent of frequency.

For example, the following is the Attenuation vs. Frequency for an RG-8 (coaxial) cable:

enter image description here

At 1Mhz the Attenuation is 0.1 dB for each 100 Feet.

At 100 Mhz is 1 dB for each 100 Feet

At 1 Ghz is around 5 dB for each 100 feet.

dB is a logarithmic measure, so these differences are huge.

Your Downstream speed (64286 kbps) uses signals in higher frequencies, maybe hundreds of Mhz.

Your Upstream speed (18782 kbps) maybe is located below 10 Mhz, so the attenuation is higher for downstream.

Here is other graphic involving multiple types of cable:

enter image description here

  • Thank you for your response. Note that my upstream speed is 18782 kbps rather than the 1872 kbps that you have quoted in your answer. The frequency ratio is about 3:1 and my house is about 750 metres from the fibre/copper cabinet. 17.4 dB seems to be a rather larger difference? – Farcher Mar 6 '18 at 9:14
  • Sorry the line length is nearer 400 metres. – Farcher Mar 6 '18 at 9:24
  • 1
    The diagram for RG-8 coax cabling may be misleading - RG-8 has a far superior performance compared to the local loop used for POTS and xDSL. The local loop's twisted pair (Cat-3ish) is at least one order of magnitude worse than RG-8. – Zac67 Mar 6 '18 at 11:37
  • For the 2nd diagram, the frequencies for fiber don't make much sense. 850 nm wavelength (for MMF) corresponds to ca. 3.5x10^14 Hz, 1550 nm (for SMF) to 1.9x10^14 Hz. 10^10 Hz is just 10 GHz which is no light at all. Also, cat 6A cable goes up to 500 MHz which is 5x10^8 Hz (beyond MMF in the diagram)... – Zac67 Mar 6 '18 at 12:52

A reported upstream attenuation of "0.0 dB" most probably means that the DSLAM does not report this figure back to the modem. 0 dB is technically not possible.

As has already been pointed out the downstream uses higher frequencies than the upstream, so it suffers more from frequency-depending attenuation. Accordingly, the downstream uses a much wider frequency band to make up for this.

Note that these figures appear to correspond to (short) copper transmission. A fiber transmission with 17 dB attenuation would indicate many dozen km in length.

  • Thank you for your answer. It is a short copper transmission from a telephone exchange and cabinet which I have been able to identify. – Farcher Mar 6 '18 at 11:54

Not the answer you're looking for? Browse other questions tagged or ask your own question.