DOCSIS 3.1 Seen Taking Off

Seeking to join the Gigabit Parade, cable operators are apparently chomping at the bit to deploy the emerging next-gen DOCSIS 3.1 spec so they can offer 1 Gig and higher speeds themselves.

In its latest survey of cable providers across the globe released earlier this week, IHS Inc. found that, on average, providers expect to pass about a third of their residential broadband subscribers with DOCSIS 3.1-enabled headends by April 2017. In the US alone, that would translate to more than 17 million cable modem homes passed by D3.1, which is designed to support data downstream speeds as high as 10 Gbit/s and upstream speeds of 1 Gbit/s or more.

If this brisk rollout pace is realized, DOCSIS 3.1 would be far more widely deployed in the early going than its predecessor, DOCSIS 3.0, as well as earlier versions of the cable broadband spec. In fact, many cable systems have still not been upgraded for DOCSIS 3.0, more than nine years after CableLabs completed the spec and more than seven years after the first MSO deployments of D3.0 began.

But, under competitive pressure from fiber-based providers that are now deploying gigabit services, many cable operators are looking to step up their game by rolling out DOCSIS 3.1 as fast and early as possible. In recent weeks, such major North American MSOs as Comcast Corp. (Nasdaq: CMCSA, CMCSK), Videotron Ltd. and Suddenlink Communications have announced plans to start deploying DOCSIS 3.1 as early as this fall. And previously such other large North American and global MSOs as Cox Communications Inc. and Liberty Global Inc. (Nasdaq: LBTY) laid out plans to upgrade their networks to DOCSIS 3.1, starting in 2016. (See Comcast Plots First DOCSIS 3.1 Trials and Videotron Catches Gigabit Fever.)

Read more about Gigabit Cities and the expansion of gig services in our Gigabit Cities section here on Light Reading.

In a closely related move, cable operators also appear eager to upgrade their access networks to a distributed access architecture (DAA) in the next two years. The IHS survey found that more than 42% of providers plan to put in place some type of DAA by 2017, with the choices split between Remote PHY, Remote MAC/PHY and Remote CCAP.

In each of the three cases, at least part of the data processing and video processing functions now conducted in the cable headend would be moved to the hybrid fiber-coaxial (HFC) optical nodes deep inside the access network. Cable technologists see DAA as a more efficient, effective way for operators to deliver higher data speeds and greater overall capacity to their subscribers than the current centralized delivery model.

"Cable operators are clearly committed to both DOCSIS 3.1 and distributed access architectures to increase bandwidth in their access networks," said Jeff Heynen, research director for broadband access and pay TV at IHS. "Though there's no consensus yet on which distributed access technology most will use, there's no question they will distribute some portion of the DOCSIS layer to their optical nodes."

Speaking of optical nodes, another new IHS report estimates that HFC optical node shipments will more than double over the next four years as cable operators install more fiber in their networks and keep splitting nodes into smaller and smaller service groups to deliver more bandwidth to customers. Specifically, the research group sees annual node shipments jumping from 92,000 units in 2014 to 200,000 units by 2019.

Presumably, optical node revenue will jump as well. IHS calculates that global node revenue reached $355 million last year, up 14% from the 2013 total. Arris Group Inc. (Nasdaq: ARRS) led the way in both node shipments and revenues last year.

Besides increasing dramatically in volume, optical node shipments will also shift dramatically in type over the next few years, IHS reckons. Whereas digital return nodes accounted for 80% of revenues and analog return nodes accounted for 15% of revenues in 2014, the research group expects the breakdown to switch to 35% Remote CCAP devices, 27% Remote PHY devices and just 23% traditional digital return nodes by 2019.

Finally, IHS sees many cable operators creating more capacity for upstream spectrum over the next couple of years, whether through mid-splits or other methods. In its survey, nearly half of cable providers said they expect to have 86 MHz to 100 MHz of upstream bandwidth available by 2017, which would be about double what they have today. And nearly a quarter said they expect to have 101 MHZ to 200 MHz of upstream spectrum by then, or as much as four times what they have now.

— Alan Breznick, Cable/Video Practice Leader, Light Reading

jabailo 7/29/2015 | 12:13:24 PM
Long term erosion When it comes to the long term operations costs for maintenance and repair, is there any competitive advantage between a cable network, a telco running optical fibers, and a telco using a mix of optical fibers and twisted pair?

Long term, it is more expensive, say, to keep a cable network alive, and growing, while fighting breakage, wear and tear, heat and so on vs. fiber?

Sign In