Market Advisory Report

HFCs Lucky Seven Technologies: How Can Cable Operators Compete with FTTH?
March 31, 2011
Issue

Erik Keith
Current Analysis Principal Analyst, Fixed Access Infrastructure

As cable operators face an increasing threat from telcos, especially those leveraging FTTx networks to deliver very competitive, advanced service bundles such as Verizon and AT&T, what options do the cable operators have that will enhance their ability to deliver competitive service bundles? Verizons GPON-based FiOS service represents the most technologically advanced competitive threat, at least in the geographic areas that have been fortunate enough to be selected for the FiOS network build out. AT&Ts U-verse service, while not delivered over a full-ber network, still leverages deep ber, i.e., ber-to-the-node (FTTN) technology that enables AT&T to oer very competitive triple-play bundles that can support multi-stream HDTV services. With this in mind, what technologies can cable operators leverage to ght back the telco threat eectively? Also, does it make sense for cable operators to continually upgrade their networks, or just transition to a full-ber access architecture more quickly?
Current Perspective

Cable operators have at least seven technologies they can leverage to defend against subscriber churn to telco operators, most of which still utilize the existing hybrid ber-coax (HFC) cable plant. Here are the Lucky Seven:

1. DOCSIS 3.0 Channel Bonding The Data over Cable Standard Interface Specication (DOCSIS) 3.0, ratied in 2006 but not widely deployed until 2009, enables cable operators to virtually bond multiple downstream and upstream channels together to deliver ultra-high bandwidth to an individual subscriber. The DOCSIS 3.0 Standard supports the following throughput capacities, listed in the table below:
Channel conguration
Number of downstream channels 4 8 Number of upstream channels 4 4

Downstream throughput
DOCSIS 171.52 (152) Mbps 343.04 (304) Mbps EuroDOCSIS 222.48 (200) Mbps 444.96 (400) Mbps

Upstream throughput
122.88 (108) Mbps 122.88 (108) Mbps

In terms of real-world service oerings, no MSOs are currently oering the 300 to 400 Mbps downstream bandwidths listed in the table, at least for residential subscribers. The highest current bandwidth oered in North America by cable operators are the 101 Mbps service (and 15 Mbps upstream) oered by Cablevision, and the 120 Mbps (and 20 Mbps upstream) oered by Quebec-based cable provider Videotron.

2011 Current Analysis Inc. All rights reserved. For more information, please call +1 703 404 9200, toll-free +1 877 787 8947 Europe +33 (0) 1 41 14 83 15. Or visit our Web site: www.currentanalysis.com

Market Advisory Report

Report:

2. CMTS Advances (Higher density CMTS blades, I-CMTS, M-CMTS) Much

like telco copper/ber access systems, increasing the density and corollary scalability of Cable Modem Termination Systems provides cable operators with additional bandwidth, for both the downstream (d/s) and upstream (u/s) paths. Cisco, ARRIS and Motorola which absolutely dominate the CMTS/cable head-end market with 99% market share have all pushed forward on the CMTS density/scalability front, enabling cable operators to deliver higher data rates to an even greater number of end users. Higher density CMTS blades/platforms in combination with DOCSIS 3.0 channel bonding have facilitated dramatic leaps forward in the bandwidth per subscriber capabilities of the operators, e.g., Cablevisions 101 Mbps d/s service in the New York metro area. Integrated CMTS (I-CMTS) and Modular CMTS (M-CMTS) solutions also provide cable operators with additional paths for capital and operational eciency gains, as well as bandwidth and service delivery eciencies. An I-CMTS combines all of the DOCSIS 3.0 channel bonding elements into a traditional CMTS system architecture (as dened by previous DOCSIS specs), while the M-CMTS architecture is dened as one where DOCSIS 3.0 channel bonding is delivered by leveraging a standard CMTS in conjunction with an Edge QAM, DOCSIS DTI timing source, and GigE interfaces. Both the I-CMTS and M-CMTS approaches oer distinct advantages (and concerns) for cable operators, depending on key factors such as service requirements (e.g., volume of on-demand trac) and network size/scalability.

HFCs Lucky Seven Technologies: How Can Cable Operators Compete with FTTH?
Fixed Access Infrastructure

3. Switched Digital Video (SDV) SDV, very simply, delivers only the requested channels to a given set-top box (STB) in the subscriber household. This is in stark contrast to more traditional non-switched video delivery method, where all of the cable operators channels are broadcast downstream into the cable network (and consequently, to the end user/subscriber household). Non-switched video delivery consumes a tremendous amount of bandwidth, while SDV consumes only a fraction of the network bandwidth. SDV does require an investment in enabling technologies, but this has proven to be a very cost-eective investment, as the same technology enables TV/video on-demand services, which most major cable operators have oered for years. Cablevision has the largest deployment of SDV-based content in North America, and it provides all of its iO International in-language packages over SDV. 4. Compression (H.264, or MPEG-4, Part 10 [Advanced Video Coding]) The transition from MPEG-2 to MPEG-4 based compression technology, like SDV, has given broadcast TV/video service providers of all stripes (i.e., cable, satellite, telco, etc.) the ability to deliver the same content, including multi-stream HDTV, utilizing less bandwidth. Specically, the transition from MPEG-2 to H.264 typically reduces the required bit rate by half, e.g., meaning that full-rate broadcast HDTV stream can be reduced from 19 Mbps to less than 10 Mbps (and of course, most operators are compressing video streams even further, especially the direct broadcast satellite (DBS) operators.
5. Increased RF Spectrum Cable operators have also been expanding their available channel capacity and bandwidth by increasing their radio frequency (RF) spectrum, i.e., up to 750 MHz and even 860 MHz, with some operators looking to go to 1 GHz and above. This provides not only the additional bandwidth headroom necessary for delivering additional broadcast TV/video channels and streams, but also additional spectrum to be leveraged for ultra-broadband data/IP trac (ironically including over-the-top [OTT] video streams, e.g., YouTube etc.). Again, Cablevision led the way on this front; Cablevisions entire 40,000 miles of hybrid ber/coaxial cable have been upgraded to at least 750 MHz, and densely populated areas have been upgraded to 860 MHz. 6. Node Splits One of the most straightforward, conceptually simple paths for cable operators to increase the bandwidth and advanced service delivery capabilities of their HFC networks dramatically is to split their nodes. In a typical cable network, all of the network trac from the

2011 Current Analysis Inc. All rights reserved. For more information, please call +1 703 404 9200, toll-free +1 877 787 8947 Europe +33 (0) 1 41 14 83 15. Or visit our Web site: www.currentanalysis.com

Market Advisory Report

Report:

HFCs Lucky Seven Technologies: How Can Cable Operators Compete with FTTH?
Fixed Access Infrastructure

cable head-end is sent downstream to multiple nodes in the network, which in turn distribute the trac to individual end users (residential subscriber households). The nodes serve anywhere from 500 to 2,000 end points. By installing new nodes in the network which are relatively adjacent to the existing nodes, cable operators can halve the number of subs per node (i.e., from 500 to 250). This of course provides an immediate doubling of the capacity per sub. The chart below, which summarizes the gains enabled by node splits, based on research conducted by leading cable infrastructure vendor Motorola, illustrates this point.

7. Radio Frequency over Glass (RFoG) RFoG is eectively a hybrid, transitional

technology implemented by replacing the HFC portion of the cable operator network with passive optical ber. In terms of the last mile technology, RFoG is very similar to passive optical network (PON), with the dierence being that upstream of the cable network node, all of the enabling technology remains the same (CMTS and other head-end equipment), and in most cases, the customer premise gear can remain unchanged (STBs and cable modems). RFoG provides multiple benets for the operator, including increased downstream RF spectrum (up to 1 GHz), increased upstream capacity, and lower operational expenditures due to the passive ber network eciencies vis--vis HFC (RFoG eliminates active elements such as ampliers, and requires far less power than coax).
Near Term Drivers

Expanded HDTV and 3D TV content will spur further network upgrades by cable operators. Even with DOCSIS 3.0, compression (H.264) and SDV, the ever-expanding array of HD and now 3D content will increasingly tax the capacity of current cable operator HFC networks. As such, operators will very likely take the next major steps in their HFC network upgrade paths, i.e., node splits and increasing their RF spectrum to 860 MHz and even 1 GHz.

2011 Current Analysis Inc. All rights reserved. For more information, please call +1 703 404 9200, toll-free +1 877 787 8947 Europe +33 (0) 1 41 14 83 15. Or visit our Web site: www.currentanalysis.com

Market Advisory Report

Report:

HFCs Lucky Seven Technologies: How Can Cable Operators Compete with FTTH?
Fixed Access Infrastructure

Data bandwidth escalation will continue, at least in hyper-competitive markets. The primary example here is the Cablevision territory where Verizon has pushed very hard with its FiOS service. Cablevisions 101 Mbps d/s service (not to mention its expanded network capacity, which will support up to 500 HDTV channels) is a very strong counterpunch to the Verizon encroachment, but Verizon can be expected to escalate this battle accordingly.
Competitor Response & Recommendations

Cable head-end vendors such as Cisco, ARRIS, Motorola and Casa Systems, need to outline their respective solution set advantages, in order to establish clear dierentiation from one another. For example, is the I-CMTS or M-CMTS architecture the best path for operators? Also, where do each of these vendors stand with respect to the development of the Converged Multiservice Access Platform (CMAP) solution? Other cable infrastructure market contenders, such as EPON (and/or RFoG) systems vendors, also need to target specic opportunities for customer and market gains. This should not be especially dicult given the evolving competitive landscape, and the corresponding (and everescalating) demand for advanced service bundles.
Buyer Actions

Cable operators competing against AT&Ts U-verse service (and/or other telcos delivering triple play services over FTTx but not FTTH networks) need ramp up their sales and marketing counter-oensives against telcos that are still utilizing copper last mile solutions. The cable operators were very successful in marketing against rst-generation DSL services (e.g., digital slow line), and should leverage the same tactic against FTTN-based services, which utilize VDSL2 and/or ADSL2x-based last mile copper connections. While U-verses per-user bandwidth is far superior to rst-gen DSL, telco copper does have limitations that cable operators can exploit to their advantage, especially when compared to upgraded HFC networks. Cable operators need to evaluate their specic service delivery requirements, based on both current competitive conditions and on their projected service roadmaps, and then select their specic service/bandwidth enabling technologies accordingly. Clearly, bandwidth demand will continue to escalate, thanks to the never-ending thirst for higher-speed data connectivity, as well as the proliferation of HD content (not to mention 3D on-demand). Cable operators also need to evaluate their options on the FTTH front, for example, does it make sense to go ahead and transition to a full ber network in the near term? If so, which specic FTTH network technology/architecture makes the most sense? Many large cable operators have already trialed EPON and 10G EPON

2011 Current Analysis Inc. All rights reserved. For more information, please call +1 703 404 9200, toll-free +1 877 787 8947 Europe +33 (0) 1 41 14 83 15. Or visit our Web site: www.currentanalysis.com