dslreports logo
 
    All Forums Hot Topics Gallery
spc

spacer

Search Topic:
uniqs
5229
share rss forum feed


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo

1 edit

1 recommendation

Technical Discussion of Upstream Channel Bonding

Split from this topic.
The original post in this topic is a reply to this post. ~sorto'


It is very unlikely that the 16QAM channel (Upstream4) is being bonded. Your modem is just detecting it. One would bond similar Channel Types as they have same modulation scheme and bandwidth.

»www.cisco.com/en/US/tech/tk86/tk ··· c3.shtml


EG
The wings of love
Premium
join:2006-11-18
Union, NJ
kudos:10

1 edit
said by EnasYorl:

It is very unlikely that the 16QAM channel (Upstream4) is being bonded. Your modem is just detecting it. One would bond similar Channel Types as they have same modulation scheme and bandwidth.

»www.cisco.com/en/US/tech/tk86/tk ··· c3.shtml

Please point out something specific in that article. My understanding has been that dissimilar channels can indeed be bonded.


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo

1 edit

1 recommendation

Well there are many reasons to not bond different types of carrier.

excerpts from my link.


And


Also there level differences when using different modulation schemes.

This article talks more about the modulation power of bonding channels. http://www.cisco.com/en/US/docs/ios/cable/configuration/guide/ubr_uscb.html#wp1181735


So in short maybe it can be done in a lab (I've yet to see anyone show it done), but I would bet money that no one bonds those different modulation schemes in the real world deployment.


netcool
Premium,VIP
join:2008-11-05
Englewood, CO
kudos:117

1 edit
said by EnasYorl:

So in short maybe it can be done in a lab (I've yet to see anyone show it done), but I would bet money that no one bonds those different modulation schemes in the real world deployment.

You would lose that bet.

Look through this thread, many many people have bonded channels using different modulation profiles.

I myself have two channels bonded, one TDMA and one ATDMA. It also helps that I can login to the CMTS to verify as well


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo

4 edits

1 recommendation

said by netcool:

said by EnasYorl:

So in short maybe it can be done in a lab (I've yet to see anyone show it done), but I would bet money that no one bonds those different modulation schemes in the real world deployment.

You would lose that bet.

Look through this thread, many many people have bonded channels using different modulation profiles.

I myself have two channels bonded, one TDMA and one ATDMA. It also helps that I can login to the CMTS to verify as well ;)

So in GW, CO you are saying your bonding 1.1 and 2.0 upstreams?

Care to show your modem?


Now looking at this data from my two upsteams what modulation is Really in use QPSK or 64 QAM? is the Modem accurate with it's data or is it false? I can tell you for a long time I had both carriers present but they were not bonded until recently as my modem is one of the few that will indicate true bonding.

Upstream Bonding Channel Value
Channel ID 4 3
Frequency 22800000 Hz 32400000 Hz
Ranging Service ID 7409 7409
Symbol Rate 5.120 Msym/sec 5.120 Msym/sec
Power Level 49 dBmV 49 dBmV
Upstream Modulation [3] QPSK
[3] 64QAM
[3] QPSK
[3] 64QAM

I will throw in the CMTS command info as well
http://www.cisco.com/en/US/docs/ios/cable/command/reference/cbl_10_cable_u_to_cable_w.html#wp1090928

Based on the vendor information I don't believe you've proven your point.

Especially when scdma-d3 is what is needed to bond.

Configures the upstream for DOCSIS 3.0 S-CDMA modulation profiles. The scdma-d3 option uses channel type 4SR mode.


qam64qam16

@scansafe.net

1 edit
Click for full size
My modem in Aurora Colorado has shown 64qam and 16qam bonded for a loooong time now, maybe over a year


gar187er
I do this for a living

join:2006-06-24
Dover, DE
kudos:5
reply to EnasYorl
up bonding does indeed make use of the 16qam.....equalization on that same 16qam is not always turned on, so nodes with noise behind it will still have issues that the modem cannot overcome.
--
I'm better than you!


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo

2 edits

1 recommendation

reply to qam64qam16
Your LED's are BLUE or Green on US indicator?

But my point before was a TDMA carrier wouldn't be bonded based on the info I posted. If you are bonding a channel that is meant for 1.1 DOCSIS modems and 3.0 modems seems like a poor design choice the Return spectrum maybe has too much noise to support multiple 64QAM channels . If both your LED's are blue DS and US, i will admit i'm blown away.

For those whom like to read more on settings in CMTS.

http://www.cisco.com/en/US/docs/ios/cable/command/reference/cbl_06_cable_m.html


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo

1 recommendation

A interesting read as well
Webview
»docs.google.com/viewer?a=v&q=cac ··· RAbDv4_Q

PPT link »www.google.com/url?sa=t&rct=j&q= ··· fxHV8Z2A


NetFixer
Bah Humbug
Premium
join:2004-06-24
The Boro
Reviews:
·Vonage
·Comcast Business..
·Cingular Wireless
reply to EnasYorl
said by EnasYorl:

It is very unlikely that the 16QAM channel (Upstream4) is being bonded. Your modem is just detecting it. One would bond similar Channel Types as they have same modulation scheme and bandwidth.

»www.cisco.com/en/US/tech/tk86/tk ··· c3.shtml

It may be "unlikely" in your opinion, but it is indeed done.



--
We can never have enough of nature.
We need to witness our own limits transgressed, and some life pasturing freely where we never wander.


netcool
Premium,VIP
join:2008-11-05
Englewood, CO
kudos:117
reply to EnasYorl
How about an example from a live uBR10012?

Expand your moderator at work


netcool
Premium,VIP
join:2008-11-05
Englewood, CO
kudos:117

1 edit
reply to EnasYorl

Re: Technical Discussion of Upstream Channel Bonding

said by EnasYorl:

Now looking at this data from my two upsteams what modulation is Really in use QPSK or 64 QAM? is the Modem accurate with it's data or is it false?

The modem is only using QPSK to transmit station maintenance data. All other data coming from the modem is using 64QAM. So technically it's using both.

See this for additional info on configuring modulation profiles:

»www.cisco.com/en/US/tech/tk86/tk ··· b0.shtml


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo

1 edit
reply to netcool
US0 fighting a ton of noise there. Ton's of Corrected and Uncorrectables


Thanks for sharing.

Reading for the geeks https://www.nanog.org/meetings/nanog46/presentations/Sunday/Byju_Intro_DOCSIS_N46.pdf

And just because you can do something doesn't mean you should.

This hits on some of my thoughts. http://www.google.com/url?sa=t&rct=j&q=throughput%20rates%20of%20qam%20&source=web&cd=4&ved=0CGYQFjAD&url=http%3A%2F%2Fpiedmontscte.org%2Fresources%2FDOCSIS_Throughput.doc&ei=8vKyT5DTCcOuiAK3mv3yAw&usg=AFQjCNES9BAzB3k2cyUTka2sabAJmIKOvw


gar187er
I do this for a living

join:2006-06-24
Dover, DE
kudos:5
doesnt matter what you think....higher engineers and telecom people are doing it with success.....i work it, troubleshoot it, and make it work....its happening.....

you have them on 1 channel.....that is not a ton of noise...so please back it up.....

i would assume the 16qam is lower in the spectrum those showing more noise due to typical 5-18 garbage.....if you work in the field you know what i mean....
--
I'm better than you!


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo
said by gar187er:

doesnt matter what you think....higher engineers and telecom people are doing it with success.....i work it, troubleshoot it, and make it work....its happening.....

you have them on 1 channel.....that is not a ton of noise...so please back it up.....

i would assume the 16qam is lower in the spectrum those showing more noise due to typical 5-18 garbage.....if you work in the field you know what i mean....


Simple math 0.1% of the codewords are being dropped on US0
5.4% of all traffic is being corrected from errors.

That is a large amount compared to US1 and US2

Not to mention that just 2% packet loss can cause robotic issues or gaming issues on UDP sessions.

Yes it's not an issue for TCP sessions.

33db SNR is not horrible but that looks like some noise in band.

JPnATL

join:2011-11-16
Bethlehem, GA

2 edits
reply to EnasYorl
Its hard to argue about something that obviously nobody is familiar with. Their is one thing that is fact not opinion and that is 16Qam is more tolerant to noise than 32 and then 64,128 and so forth.
Expand your moderator at work


EnasYorl
Thieves World

join:2001-12-02
West
Reviews:
·Anveo
reply to JPnATL

Re: Technical Discussion of Upstream Channel Bonding

said by JPnATL:

Its hard to argue about something that obviously nobody is familiar with. Their is one thing that is fact not opinion and that is 16Qam is more tolerant to noise than 32 and then 64,128 and so forth.

This is true and the TDMA carrier listed is the 16QAM the other two are DOCSIS2.0/3.0 most likely 32 or 64QAM The issue I referred to is on the US0 ATDMA carrier.