DOCSIS 3.1 has finally arrived! After years of talk and development, we are finally beginning to see the initial roll-outs of this next-generation technology. A few years have passed since the benefits of DOCSIS 3.1 were touted.
Are those benefits still relevant today? It's worth revisiting.
In order to fully understand the benefits of DOCSIS 3.1, it is necessary to understand the boundaries of DOCSIS 3.0. DOCSIS 3.0 was a transformational technology in its own right and time. It provides the capacity to provide up to 1 GB of data to a service group, the ability to offer a high class of service, and provides many features and functionality that help operators with managing the customer, reporting, and reliability.
Over time, the strengths of DOCSIS 3.0 become its weakness. The ability to achieve 1 GB downstream with up to 32 QAM becomes a limitation. Long-term bandwidth projections predict that DOCSIS 3.0 will begin to reach maximum capacities as soon as 2019 (without continuing to scale down service group sizes). In addition, competition driving 1 GB classes of service has accelerated the need for something beyond DOCSIS 3.0. The once high service group capacities of 3.0 platforms are now no longer enough. As service groups migrate to smaller and smaller groups of homes passed to manage bandwidth availability, more and more ports are required. The continual scaling of chassis, optics and other equipment to accommodate this growth becomes unsustainable.
To put this scale into perspective, some operators have said they will need to split nodes from 4 to 10x what they are today over the next 10 years. And this is with the full capacity used on 3.0 chassis. The result would be 10x the CMTS chassis, 10x the optics, and 10x the nodes. Facilities, rack space and power requirements cannot scale with this growth.
For a time, these inevitabilities were pushing many operators to consider a wholesale infrastructure transition to FTTH and PON technologies. The challenge with this was the complete overhaul of the entire network from video, to data provisioning, to OSP cabling and equipment to CPE. The cost, technology and knowledge change and disruption to the customers (and roadsides) made this a very unattractive option.
Enter DOCSIS 3.1. The first problem solved is the 32 channel limitation. DOCSIS 3.1 provides the ability to bond much larger groups of spectrum together to provide a true 1 GB Class of service and beyond. This also assists in the scaling problem. Whereas before, node segmentation would often be required when groups meet the 32 QAM limitation; the ability to use the full spectrum for data removes that requirement.
DOCSIS 3.1 also allows for enhanced spectral efficiency. For math purposes, consider that a 3.0 256 QAM channel provides approximately 40 MB of throughput. DOCSIS 3.1 uses Orthogonal Frequency Division Multiplexing (OFDM) technology that allows QAM modulations to reach 1, 2, 4k and beyond. A 1k QAM provides approximately 50 MB of throughput or a 25% increase in the same amount of spectrum. When you combine this capability with distributed access architectures (DAA), we see added improvement resulting in 4k QAM modulation and beyond. Therefore, DOCSIS 3.1 provides the bandwidth with more ‘bang for its buck’.
High level-comparison features and capabilities of Next-Gen 3.1 platforms vs legacy 3.0. *Numbers may vary slightly by vendor chassis
Originally, the new DOCSIS 3.1 and DAA technologies were designed with smaller and smaller cascades in mind. However, testing has shown that improvements can be made over some of the longer cascades that exist today. For example, it is possible to achieve 1024 QAM where 256 QAM currently exists. This improved performance continues to increase as you get down to smaller and smaller cascades.
Addressing the Upstream
As data rates increase, the upstream continues to become more and more of a choke point. Studies suggest that the upstream capacity should be 10% of the highest class of service offered. For example, for a 1 GB service to be fully functional, approximately ~100MB of upstream throughput is required. As larger and larger data pipes are brought to each service group, the upstream limits will be pushed. DOCSIS 3.0 allows for a 5-85 upstream, allowing room for growth to handle this change. DOCSIS 3.1 pushes the split to 5-200 which allows for HFC systems to theoretically achieve a GB symmetrical service.
The Importance of DAA
DAA architectures such as Remote Phy or Remote Mac/Phy are inseparable from DOCSIS 3.1 when discussing the benefits next-gen DOCSIS platforms. While 3.1 chassis do traditionally offer a higher amount of port density in a chassis, this still becomes a limitation of the box. The next generation of CCAP platforms have more throughput potential than the physical RF output limitations can take advantage of. DAA becomes extremely valuable in that it removes that limitation by providing a digital link to the node itself, eliminating the limitation of physical RF ports. This also provides better link performance which continues to compliment the ability to achieve higher orders of modulation (better throughput performance across the same amount of spectrum).
|Improvement in MB throughput of spectrum by leveraging higher orders of modulation made possible by DOCSIS 3.1 and DAA.|
Perhaps the greatest benefit of DOCSIS 3.1 is that it dramatically extends the life of the HFC network and physical architecture. By extending the life of the physical infrastructure, it extends the life of all the assets of the network—from the video platform, existing CMTS chassis and provisioning systems, optical infrastructure, OSP, and CPE.
The new urgency of a long-term plan
DOCSIS 3.1 in many ways did swoop in and save the day, but it also brings to light a flaw and errors that cannot be made again. For nearly a decade, many cable operators got trapped in operational mode without a long-term strategy. Had 3.1 not come along, the push to get to FTTH would be exploding at a rate that the supply cannot provide. 3.1 has brought new life to existing infrastructure and has allowed for a more graceful migration to fiber deep, higher bandwidth capacities, system upgrades, service migration and virtualization. All of these solutions need to be executed with an eye on the longer term future, to ensure that the things we do today compliment the needs of tomorrow instead of simply extending the limits of the past.