dslreports logo
 
    All Forums Hot Topics Gallery
spc
uniqs
112

telcodad
MVM
join:2011-09-16
Lincroft, NJ

telcodad to DocDrew

MVM

to DocDrew

Re: 8 downstream 8 upstream modem?

said by DocDrew:

Cable ops would have to change out all the diplex filters in their gear to get more upstream spectrum, but then it would impede on the lower broadcast channels (channel 2 starts at 50 Mhz).

Yes, ARRIS demonstrated a "high-split DOCSIS upstream system" at the 2011 Cable show in Chicago: »www.prnewswire.com/news- ··· 744.html

"The demo will also show a proof-of-concept implementation of a 5-200 MHz high-split DOCSIS upstream system with 575 Mbps of DOCSIS upstream bandwidth being transmitted over 24 DOCSIS Upstream channels out of a single Fiber Node. This demo uses a single ARRIS 24U CAM to receive the high-split 5-200 MHz upstream spectrum from a Fiber Node. This record-breaking upstream bandwidth is intended to help the industry explore the future frequency allocations of HFC to protect its value and viability for many years to come."

I also found this item from a year ago on the CED Magazine website:

Cox, Motorola lay claim to new return path speed record
CED Magazine - March 1, 2011
»www.cedmagazine.com/news ··· d-record

"Using a Motorola DOCSIS 3.0 RX48 return path receiver module operating within a Motorola BSR 64000 cable modem termination system (CMTS) edge router, the record of 356 Mbps for a 5-85 MHz return path was set by transmitting across 12 return path channels, of which six channels employed 256 QAM modulation. Motorola said typical HFC networks today have two or three upstream channels delivering an aggregate of 40-70 Mbps.
:
In addition to establishing the world record at 5-85 MHz, a maximum transmission rate of 141 Mbps was also achieved over a 5-42 MHz return path using six return path channels. Three of these channels were able to operate using 256 QAM modulation, as opposed to 64 QAM maximum in use today, which is also believed to be a record.

Motorola said that for the first time, cable operators could use the 5-42 return path to provide 100 Mbps dedicated to business customers, while at the same time providing residential customers with the return path bandwidth necessary to meet their growing needs."
telcodad

telcodad

MVM

said by telcodad:

said by DocDrew:

Cable ops would have to change out all the diplex filters in their gear to get more upstream spectrum, but then it would impede on the lower broadcast channels (channel 2 starts at 50 Mhz).

I also found this item from a year ago on the CED Magazine website:

Cox, Motorola lay claim to new return path speed record
CED Magazine - March 1, 2011
»www.cedmagazine.com/news ··· d-record

:
In addition to establishing the world record at 5-85 MHz, a maximum transmission rate of 141 Mbps was also achieved over a 5-42 MHz return path using six return path channels. Three of these channels were able to operate using 256 QAM modulation, as opposed to 64 QAM maximum in use today, which is also believed to be a record.

Motorola said that for the first time, cable operators could use the 5-42 return path to provide 100 Mbps dedicated to business customers, while at the same time providing residential customers with the return path bandwidth necessary to meet their growing needs."

BTW - I found this thread from April 2011 on the possible use of (and problems with) higher-order QAM carriers (256-QAM on the upstream, 512-QAM and 1024-QAM on the downstream) to increase DOCSIS transmission rates:

Forums>US Cable Support>Cable users>QAM 512/1024
»QAM 512/1024

"Well Ron Hranac said:

1024-QAM requires a very clean and tight plant, and about 6 dB higher carrier-to-noise ratio to achieve the same bit error rate as 256-QAM (for instance, 1.0E-08 BER requires approx. 34~35 dB CNR for 256-QAM; the same BER with 1024-QAM needs around 40~41 dB). As well, 1024-QAM is very susceptible to even small amounts of phase noise, requiring new tuner designs in set-tops, cable modems, and the like. Depending upon the plant's performance, it likely will require increasing the digital channel power of 1024-QAM signals to overcome the CNR requirements, risking downstream laser clipping."
telcodad

1 edit

telcodad

MVM

said by telcodad:

BTW - I found this thread from April 2011 on the possible use of (and problems with) higher-order QAM carriers (256-QAM on the upstream, 512-QAM and 1024-QAM on the downstream) to increase DOCSIS transmission rates:

Forums>US Cable Support>Cable users>QAM 512/1024
»QAM 512/1024

"Well Ron Hranac said:

1024-QAM requires a very clean and tight plant, and about 6 dB higher carrier-to-noise ratio to achieve the same bit error rate as 256-QAM (for instance, 1.0E-08 BER requires approx. 34~35 dB CNR for 256-QAM; the same BER with 1024-QAM needs around 40~41 dB). As well, 1024-QAM is very susceptible to even small amounts of phase noise, requiring new tuner designs in set-tops, cable modems, and the like. Depending upon the plant's performance, it likely will require increasing the digital channel power of 1024-QAM signals to overcome the CNR requirements, risking downstream laser clipping."

For those of you who can handle the more "technical stuff," I found this excellent presentation by Cisco's Ron Hranac on return path (upstream) design, characteristics and troubleshooting:
»www.ciscoknowledgenetwor ··· ting.pdf
telcodad

telcodad

MVM

At the 2012 Cable Show this week, there were a number of technical sessions that discussed some possible enhancements to and the future evolution of the current cable network (e.g., “HFC 2.0: The Evolution of Architecture” and “The Gigabit Network: New Possibilities in HFC.”).

Some ideas presented were - expanding to frequencies up to 1.5 GHz and beyond, going to a mid- or high-split architecture to increase upstream capacity, moving to higher-order modulations like 1024-QAM and using new, more efficient FEC schemes, such as low-density parity check (LDPC) codes.

You can read more about this in an article on the CED website:

@ The Cable Show: Hey, do you remember the future?
By Brian Santo, CED Magazine - May 23, 2012
»www.cedmagazine.com/news ··· e-future

JigglyWiggly
join:2009-07-12
Pleasanton, CA

JigglyWiggly

Member

why do you guys even care about ipv6?
only asian countries need to worry about being subnetted

DocDrew
How can I help?
Premium Member
join:2009-01-28
SoCal
Ubee E31U2V1
Technicolor TC4400
Linksys EA6900

1 recommendation

DocDrew

Premium Member

said by JigglyWiggly:

why do you guys even care about ipv6?
only asian countries need to worry about being subnetted

Huh? every IP is subnetted. Or did you mean NAT'd?

Either way do you realize that Comcast alone ran out of private IP space it uses for modem and cable box management years ago? It's pushing for IPv6 usage quicker than most (especially internally) just for that reason. They had to start using public IPv4 address space just for CPE management.

Besides that IPsec, QoS, and Multicast support is mandatory in IPv6 products, so end to end (across the internet, not just within a providers network) IPsec, QoS, and Multicast usage becomes much more feasible and common.

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

EG

Premium Member

said by DocDrew:

said by JigglyWiggly:

why do you guys even care about ipv6?
only asian countries need to worry about being subnetted

Either way do you realize that Comcast alone ran out of private IP space it uses for modem and cable box management years ago? It's pushing for IPv6 usage quicker than most (especially internally) just for that reason. They had to start using public IPv4 address space just for CPE management.

The good Dr. is absolutely correct !