CABLE NEXT-GEN EUROPE DIGITAL SYMPOSIUM – There's no pressing need for the bandwidth-expanding and upstream-boosting DOCSIS 4.0 platform, but operators are already thinking about how they can prepare the network with new amps and taps for future upgrades as other new network and home-side products remain in development.

CableLabs completed the DOCSIS 4.0 specs in March, paving the way for a possible upgrade of HFC networks to 1.8GHz spectrum, the targeting of a 10 Gbit/s downstream and 6 Gbit/s upstream, and the inclusion of new low-latency and security features and technologies.

While it will be some time before headend and consumer premises equipment (CPE) for DOCSIS 4.0 enter the scene, operators expect to have access to other types of network devices that can be deployed ahead of possible upgrades to a 1.8GHz spectrum ceiling, a sizable step-up from today's networks that are deployed at 750MHz, 860GHz or even 1.2GHz.

As part of that coming buildout, Cox Communications expects to completie the first tests on new 1.8GHz passives "fairly soon" and start deployments by the end of 2020 and into 2021," Jeff Finkelstein, executive director, advanced technologies at Cox, said yesterday during a session focused on DOCSIS 4.0 developments.

Cox, he said, will start testing DOCSIS 4.0-compatible active devices as soon as they are ready to enter the lab (perhaps later this year and into early 2021), followed by possible deployments in the summer or fall of 2021. Meanwhile, Cox will also continue to work with vendors on their timelines for other types of equipment that will make up the DOCSIS 4.0 network, including remote PHY and remote MAC/PHY devices.

"We do have a good runway with DOCSIS 3.1, but this stuff takes time to deploy," Finkelstein said. "The sooner we can stage those [devices] and get them into our planning cycle and into our supply chain, the easier it becomes on the backend."

Preparing for tomorrow... today
Finkelstein noted that peak traffic levels on residential cable networks have surged during the COVID-19 pandemic and that utilization is expected to stay above pre-pandemic levels even as people start to return to work. The cable industry, he said, needs to prepare its next network moves before DOCSIS 4.0 is truly required.

"This is our opportunity to build the highways … and get the plant to 1.8[GHz]," Finkelstein said. "This is the time for us to get ahead of that curve…This is our chance to really prepare for that next step that we have to take."

In the Netherlands, VodafoneZiggo is still in the midst of a DOCSIS 3.1 upgrade and doesn't anticipate needing DOCSIS 4.0 for about five more years.

"There is no direct necessity right now in the Netherlands to move it from our current DOCSIS 3.1 deployment into a quick DOCSIS 4.0 deployment," Leo-Geert Van Den Berg, director of fixed technology at VodafoneZiggo, said.

The current shift to 1.2GHz will provide a good stepping stone to get to 1.8GHz at a later date. "It will get there and we will go there, but it will take a number of years," Van Den Berg said of VodafoneZiggo's expected need for D4.0.

Hanno Narjus, senior VP at Teleste, agrees there is no imminent sense of urgency for cable operators to adopt DOCSIS 4.0, particularly in Europe (save for countries such as Spain and Germany), where some operators have already upgraded their plant to 1.2GHz and have upstream spectrum built out to 85MHz or even as high as 200MHz. (By comparison, most US cable operators still have upstream built out to just 45MHz.)

"I think the big momentum and the sense of urgency for 4.0 will be in the US," Narjus said. In the meantime, Teleste is now developing DOCSIS 4.0 products, including already-launched passives and amps that are currently in the works. The bigger question mark involves nodes for distributed access architectures and the availability of silicon for those products, he said.

In a separate presentation, Jeff Heynen, senior research director, broadband access and home networking at Dell'Oro Group, said he likewise is less bullish about DOCSIS 4.0 taking hold in Europe in a big way before it does in North America.

"There is enough headway with DOCSIS 3.1 and high-split architectures to allow [European cable operators] to get to near-symmetric or even symmetric speeds within the D3.1 framework that will allow these operators to compete," Heynen said.

That could change, he added, if new subsidies or private equity emerge that plow money into FTTP networks in Europe in the wake of the pandemic and cause cable operators in the region to push up their DOCSIS 4.0 plans.

Revisiting the FDX vs. ESD debate
The fact that there's no looming need for DOCSIS 4.0 could be considered a fortuitous situation given the current state of product development and the uncertainty that some cable network and CPE suppliers are facing.

The specs support both Full Duplex DOCSIS (FDX), a technique favored almost solely by Comcast, and Extended Spectrum DOCSIS (ESD), so device makers and silicon providers will likely seek out more stringent guidance from operators on what they want and need before making big R&D investments.

"That compromise solution doesn't provide a clear path forward" for vendors and chipmakers, Heynen said.

Heynen agrees that planning for a move to 1.2GHz and 1.8GHz ahead of DOCSIS 4.0 deployments will drive purchases of amps and taps. But he stressed that the uncertainty created by the FDX and ESD debate could cause timelines for DOCSIS 4.0 to slip.

Doug Jones, principal architect at CableLabs, said the specs allow vendors to build DOCSIS 4.0 products that support either ESD or FDX rather than requiring them to support both in the same product. "That's a product choice," he said.

In the case of amps, they will need to be purpose-built for ESD or FDX, as it's not possible to do both in the same design, Narjus said.

Although FDX is more challenging than ESD from a cost and operational standpoint, there are definite use cases for both technologies, Finkelstein said.

FDX requires a "node+0" architecture that pushes fiber deeper into the access network and eliminates the amplifiers between the node and the premises and puts the forward and reverse path traffic on the same band of spectrum. Finkelstein suggested that it could be attractive for multiple dwelling units and other areas with "close clusters."

He views ESD as a more general-use technology, since it relies on the same sort of Frequency Division Duplex technology in use today on DOCSIS networks to split and separate the spectrum for upstream and downstream traffic. "It's not rocket science; our techs know how to do it," Finkelstein said of ESD.

Finkelstein also believes that the discussion about DOCSIS 4.0 should be framed more holistically to include the other features, such as low latency and enhanced security, and not get bogged down in the ESD/FDX kerfuffle.

There are also other things to consider as operators mull DOCSIS 4.0 and a push to 1.8GHz. Operators, for example, will need to maintain a clean plant and be wary of leakage and ingress, with the possibility of LTE traffic creeping in, noted Chris Bastian, CTO and SVP of engineering at SCTE/ISBE.

DOCSIS 4.0 is in its infancy, but the industry is already exploring new technologies that could support spectrum in the neighborhood of 3GHz.

"Coax still has more fuel in the tank even after what we do with DOCSIS 4.0," Jones said. "We think there's even another possible turn of the crank here once we get past DOCSIS 4.0."

Related posts:

— Jeff Baumgartner, Senior Editor, Light Reading