Digging Deeply Into DOCSIS 4.0

Building on the earlier versions of the technology from CableLabs, DOCSIS 4.0 promises to bring downstream speeds as high as 10 Gbit/s and upstream speeds as high as 5 Gbit/s to the cable HFC plant. That amounts to roughly 200 times the amount of spectrum that the original, single-channel DOCSIS 1.0 specs supported a little more than two decades ago, according to Doug Jones, a principal architect at CableLabs who is shepgerding the new spec.

Moreover, DOCSIS 4.0 will bring lower latency levels and higher levels of security to cable's HFC networks, enabling operators to offer even more advanced services such as multi-player online gaming, virtual and altered reality, 8K video and more to their customers. Thus, expectations are high for DOCSIS 4.0, with the specification ratified and reports that the first devices may arrive within a year.

We spoke with Juan Bravo, VP of sales for cable equipment maker Antronix, about what the next-gen DOCSIS spec will mean for the cable industry.

LR: From Antronix’s point of view, what are the key changes DOCSIS 4.0 will require for cable HFC networks, particularly when it comes to the “last-foot” connection to the customer?

Bravo: The jury’s still out on the question of whether the newly available spectrum of DOCSIS 4.0 will be brought into the home or premise, so we’ll have to wait and see. But under any circumstances its very likely that the premise architecture will require more specialized passive devices to take advantage of this environment, while maintaining as much of the legacy devices and cabling infrastructure as possible. It’s a tremendous amount of work and time to touch every home in the network to upgrade performance.

So the technology and designs will need to provide an economical and reliable way to transition without causing issues due to signal conflicts with MoCA channels above 1218 MHz, for example, and the ever-increasing presence of Wi-Fi as a home transport. The good news is that the bandwidth constraints that operators are hoping to mitigate with the new spectrum are primarily in the access plant – the “big pipe,” so to speak – and home network communications should be well-suited for high-speed communications already. Provided the back end of the network speed and capacity are keeping pace with demand, the home networks will be capable of supporting it.

Antronix has already introduced 1.8 GHz passives for the home network, and we will continue to expand our drop passive portfolio to evolve along with the implementation of DOCSIS 4.0 in the premise network. This will add to an already significant line of home devices that meet and exceed these criteria and are highly adaptable, while also optimizing performance for next-gen applications. These will include a range of hybrid devices that will deliver excellent performance and optimally differentiate the premise from the access plant for best outcomes. They will also provide more robust security against signal proliferation as output levels increase with frequency and performance parameters – becoming more stringent than they have ever been. Our deep experience and practical know-how in RF device design will help the industry capitalize on the potential of these new architectures.

LR: Some cable operators like Comcast are looking at the Full Duplex DOCSIS part of DOCSIS 4.0 to boost upstream bandwidth. But it’s a much different signaling scheme compared to earlier DOCSIS iterations, so what will that require in terms of plant upgrades and modifications?

Bravo: Although we’re not expecting a large-scale deployment of FDX technology in the very near term, we are keenly focused on whether our products can support its application in whatever scale and timing operators will need. Multi-Taps, for example, will require some specialized performance so that the FDX service groups can perform to expectations and deliver on the promise of nearer-to- symmetrical upstream and downstream performance that could be possible with FDX.

For example, our next-gen line of Milenium Multi-Taps, which will provide the industry’s only retrofit, no-splice solution for existing terrestrial architectures, will be optimized for FDX performance and can potentially increase the number of users an FDX network can support.

LR: Moving to 1.8 GHz under the Extended Spectrum DOCSIS part of DOCSIS 4.0 will help cable operators offer higher-bandwidth services. But it will present challenges too, including signal attenuation over coaxial cabling. How can these issues be addressed?

Bravo: Network interfaces will be more important than ever, and even dispositive in determining whether a given network can achieve the promise of Extended Spectrum DOCSIS and whether the gains in throughput at end-of-line are significant or even worthwhile, given all the effort to upgrade. It all starts at the tap – which is both the critical enabling factor in a frequency expansion (but potentially its weakest link) and the primary gating factor in a successful upgrade.

As with all higher-frequency applications, the performance parameters are tighter than they ever have been. For us, this means paying special attention to the interfaces in our passive products and characterizing our actives for performance in an Extended Spectrum deployment. Signal loss is, of course, a critical issue, but there are more complexities than before with higher output power as a factor in the design of a next generation tap cascade.

Traditionally this was addressed through additional amplification, but this also exacerbates noise and can degrade MER significantly, especially in the upper range of the spectrum at end-of-line and can impact the viability of higher-order modulation.

One solution being looked at is the introduction of low-power and low-gain inter-stage amplifiers that could improve end-of-line with less impact on MER. This area of the network where these architectural modifications and innovation are taking place has been our natural environment since our inception. So it’s an area we know very well and expect to have significant presence across a variety of solutions that are now in development and timed for release consistent with the expected upgrade cycle and availability of gain hybrids in a commercially complete package of line amplifiers, as well as CPE silicon that will ultimately drive the deployment numbers quickly to follow passive and active network components in affordable and highly reproduceable designs.

At Antronix, we’re keenly focused on designs that meet cost-benefit evaluations, but we also consider that not all operators or networks are the same, so we are always committed to providing the best of a variety of options. Our Extended Spectrum Multi-Tap roadmap is a great example of that approach. Operators that have extensive deployments of our flagship Milenium platform will be able to very cost-effectively increase the spectrum of their passive networks, adding in many cases up to 50% to their capacity, by simply swapping out a face plate, saving significant cost and, most importantly, causing the least disruption to downstream subscribers. We are evaluating these incremental options closely for cost benefits and the potential for working with DOCSIS 4.0 hybrids.

From a customer satisfaction and time-to-market perspective, this could be an invaluable benefit. These MGT-G6 series taps will be an excellent option for a vast number of networks, and they will be available cost-effectively soon. Oher areas of the network, due to market demographics, will be more viable for a full-spectrum upgrade in the passive capacity to 1794 MHz or above, and a new housing which, though requiring a re-splice, will allow future upgrades up to 3 GHz. Antronix also has a solution ready for those scenarios, and we’ll be releasing that product – our MTH line – right on the heels of our retrofit solution.

LR: Getting fiber closer to the customer also has been a trend, opening the door for software-defined technologies such as Distributed Access Architecture (DAA). But it isn’t as simple as replacing coax with fiber. What are the elements cable operators need to be mindful of?

Bravo: No, it’s certainly not as simple as replacing coax with fiber. DAA will require a transition to digital-forward transport for cable operators and systems interoperability with existing back office – a significant move away from proprietary systems designed to make DAA more prevalent and benefit the economies of adopting the solution comprehensively. Also, DAA does not yet directly address the need for more fiber density in the long-haul transport, which is a challenge pretty much everywhere.

So new solutions will also need to emerge, such as coherent optics, node aggregation and maybe others that begin naturally to address how potentially a large majority of all nodes in service today can be capable of delivering upwards of 10 Gbps. The capacity just isn’t there now to do that at scale, but it will need to be. At Antronix, we’re looking very closely at how the DAA model can be both simplified and expanded deeper into the network. The increasing prevalence of more and more sophisticated actives in the cable network, I think, has been a little bit of a surprise development that cable operators now face – after years of moving aggressively to remove the number of actives in the cable plant and improve network reliability and uptime, while reducing power utilization.

As a result, the next wave of devices will need to be much more robust and improve upon the ease of implementation and interoperability that naturally has been one of the features of a fully realized DAA deployment. In particular, our xPON Hybrid Taps will provide yet another option to reach gigabit symmetrical speeds, in an approach that combines the benefits of digital optics with seamless interoperability even with existing DOCSIS networks. That’s our Lancet Tap solution. It is extremely well-timed for introduction at this point in the application of digital and the increasing depth of fiber penetration. It is the future of DAA in a DOCSIS 4.0 network.

The fiber architecture required for an eHFC FTTT deployment is also very consistent with the ongoing evolution of HFC and will not be lost as operators driver fiber deeper into the plant. Lancet will provide a very cost effective incremental and fully backwards compatible alternative for operators who are committed to deeper fiber in a methodical and investment-savvy approach. One that is not as heavily front-loaded as FTTH – largely unviable for cable in most markets for that reason. Lancet will allow operators who have already invested in a PON backbone to moderate their deployment intelligently and have more control over the pace of deployment. It allows all of the upgrades to occur with no downstream disruption to subscribers, almost entirely within the operator's demarcation, and with highly predictable rollout and excellent time-to-market characteristics.

LR: At the same time, going to fiber-to-the-premises (FTTP) networking may be cost-prohibitive in all portions of the cable network. What options do cable providers have to boost capacity using existing coax connections to the customer?

Bravo: DAA will go a long way towards addressing that challenge provided the backbone capacity can be greatly expanded. But eventually we believe the PHY layer could move even closer to the subscriber premise, enabling far better scalability than legacy centralized architectures. The distribution point holds a lot of promise to accommodate more economical designs that can be more viably deployed closer to the subscriber, making the economics of running that additional length of fiber – especially given the long-term operational benefits – perhaps much more attractive than it has been to date.

That traditional outlook may be changing in positive ways. Our Intercept eHFC technology team has been focused on this question of scale and the relative benefits of fiber-deeper architectures from the outset. We’ve been looking at the economics of fiber and cost respective of benefit for a hybrid FTTT architecture in particular. Lancet will provide a very effective alternative to dropping fiber at the premise and will still deliver PON performance, as well as aggregate all existing DOCSIS data capacity – all in a very straightforward, interoperable design and system that we believe could alleviate so many of the economic barriers faced by operators in conventional FTTH builds. We believe, whether aerial or underground, eliminating the need to run fiber to every home has incredible benefits. If you can then also keep all the benefits of all of your existing coaxial infrastructure, that’s clearly beneficial. It also has the additional benefit of incrementally extending the fiber plant on an investment curve that’s much more palatable from an ROI perspective.

As I’ve said, and as Antronix has shown in its engineering and manufacturing, we are deeply committed to making the most of the HFC plant. We are evolving as it evolves, accessing its potential to provide the most value for our customers.

— Juan Bravo, Vice President, Sales, Antronix

This content is sponsored by Antronix.

Be the first to post a comment regarding this story.
Sign In