MSOs sift through the 'RFoG' at SCTE's Cable-Tec Expo

by Meghan Fuller Hanna

The cable multiple-systems operators (MSOs) are in a somewhat precarious position. The telcos have been aggressively pursuing network upgrades -- in Verizon's case, via FTTH and in AT&T's, mostly via FTTN -- to deliver the requisite bandwidth to support voice, data, and video services. Therefore MSOs must respond to this strategic and tactical threat by bolstering their own offerings.

At the recent Society of Cable Telecommunications Engineers (SCTE) Cable-Tec Expo in Philadelphia, independent consultant Victor Blake (www.victorblake.com) delivered a presentation entitled, "Cable's competitive response to Verizon FiOS," during which he argued that the cable operators' entry into the FTTH world is no longer an "if" but a "when." He noted that Verizon has publicly announced plans to enhance its system to accommodate 40G upstream and 10G downstream. Assuming Verizon retains its currently favored 1x32 split ratio, it will be able to deliver 40Gx32 subscribers, or more than 1 Gbit per subscriber. And when you add a WDM-PON overlay, said Blake, it gets even worse -- for the MSO competitor, that is.

For his part, Blake believes that "FTTH is more capable than any known or currently foreseeable coax technology," and he maintained that MSOs "can easily offer any PON-based services." But he also cautioned that the MSOs need to get the timing right -- too early and they tie up precious capital, too late and they may miss the opportunity to compete. "It's not a crisis now," he said, "but we can't wait until it becomes a crisis either."

That said, most who wandered the halls of the Cable-Tec Expo would likely admit that the situation is not as simple as, "Hey, let's all go out and deploy FTTH tomorrow." In reality, the cable operators -- particularly the larger, publicly owned entities -- are hampered by Wall Street's revenue expectations. They spent billions of dollars on their HFC networks in the 1990s and simply cannot justify a massive rip-out-and-redo initiative like Verizon's.

So how can the MSOs convince their customers they have a competitive triple-play offering while also nullifying Wall Street analysts' fear of fiber costs? The most likely scenario is that the MSOs will adopt an FTTH-like technology, but one that leverages their existing architectures and back office systems, thereby mitigating some of the expense. And, in an attempt to disassociate themselves from the telcos' capital-intensive FTTH projects, the MSOs are devising their own roster of terms, including RF over glass (RFoG) and DOCSIS over PON, to describe these cable-centric FTTH technologies.

According to Floyd Wagoner, senior manager of marketing in Motorola's (www.motorola.com) Access Network Division, the cable MSOs have two priorities when it comes to PON: 1) They want a seamless back office implementation, and 2) they want to incorporate this technology into their existing HFC networks.

Both RFoG and DOCSIS over PON enable the cable operators to leverage their existing infrastructure, including their set-top boxes and DOCSIS modems. They also allow them to reap the benefits of the newly developed DOCSIS 3.0 specifications, which feature a channel bonding technique to increase the capacity of the HFC network.

The equipment vendors were out en masse at the Cable-Tec Expo, touting both RFoG systems (including Alloptic, CommScope, Motorola, Hitachi, Calix, Aurora Networks, and Tellabs), and some form of DOCSIS over PON (including Motorola, Cisco, and Salira Systems).

If recent activity among SCTE members is any indication, the equipment vendors can feel reasonably assured that they are chasing a viable market. Last fall the organization established the RFoG work program and assigned the program's projects to the SCTE Interface Practices Subcommittee (IPS) for development as an industry standard. Shane Eleniak, vice president of marketing and business development at Alloptic (www.alloptic.com), attended the most recent meeting, held concurrently with the Cable-Tec Expo, and says the subcommittee has devised a working draft of the standard. Now it's working through the optical and RF characteristics. According to Eleniak, there has been "a lot of consensus in the industry between the MSOs themselves and the different equipment vendors on what [the standard] should look like."

For example, he says members are starting to agree on which wavelengths should be used in the RFoG implementation. "What we see is 1,550 nm down and either 1,310 nm or 1,590 nm in the upstream," he explains. "And that's really being driven by, 'Do they want to use that same fiber for other things, like PON-based commercial services?' In that case, they would use the 1,590-nm return path. If there's no concern with that, then they would use the 1,310-nm return path."

Eleniak says that subcommittee is about two or three meetings away from a final standard. Given that the subcommittee meets every three months, a standard could be in the offing early next year.

Timing must be right

As Blake noted in his Cable-Tec presentation, the MSOs must time their migration to FTTH correctly. The consensus on the show floor seemed to be that the MSOs are still some years away from widespread deployment of FTTH in residential environments. On the residential front, they are mostly likely to begin with an RFoG or DOCSIS-over-PON deployment in greenfield environments where the developer has put out an RFP for FTTH.

Carl Meyerhoefer, vice president of marketing and business development at CommScope (www.commscope.com), reports that some MSOs are considering FTTH-like architectures in rural areas as well. Many MSOs have refrained from offering service in rural areas because an HFC network is not well suited to such applications. Longer-distance HFC spans require additional active components, such as amplifiers, which add to the capital cost of reaching those consumers.

"In rural applications where homes-per-mile density is fairly low, you can actually build a fiber-to-the-home network less expensively than you can an HFC network," contends Meyerhoefer, who notes that smaller MSOs tend to operate in rural areas, though some of the larger ones have rural properties as well.

Several MSOs have already deployed PON systems for commercial services and see this as a good way to start investing in fiber. As one industry insider notes, "Wall Street doesn't have a problem when they say they're dropping a fiber to a commercial customer." In this instance, MSOs may even use so-called "telco-grade" PON systems. On the commercial services front, they actually want to have more of a telco-mentality; enterprise customers are unlikely to care whether their services are deployed via DOCSIS or not.

Both Calix and Motorola announced at the Cable-Tec Expo that their GPON systems have been deployed by MSOs for the delivery of commercial services. Ontario-based Compton Communications selected Motorola's Cable PON portfolio to deliver services to commercial customers who require greater bandwidth than the traditional HFC network can deliver. And Calix inked a deal with WOW! Internet, Cable and Phone, the 11th largest cable MSO in the U.S.

Looking ahead

The Cable-Tec Expo underscored the MSOs' need to bolster the bandwidth capabilities of their HFC networks and highlighted the feverish activity among PON vendors to develop systems tailored to their specific needs. While there was much talk about EPON versus GPON, there does not appear to be industry consensus around one flavor of PON versus the other.

Vendors on both sides are hyping their own strengths and the other's weaknesses, but some wonder whether this is ultimately an empty argument. As Blake noted in his presentation, there is a strong push to converge EPON and GPON in the future. Market forces may motivate chip, component, and system vendors to produce a single platform -- one that is hopefully backward compatible with both EPON and GPON. By the time the cable operators are ready to deploy FTTH in nongreenfield residential environments there may be a new architectural option: some next-generation PON technology likely based on 10G.

However, while the IEEE and ITU have explored this possibility, one industry insider confirms that the two standards bodies have yet to even agree on which wavelengths to use, making any forward momentum unlikely for the foreseeable future. For now, then, vendors will continue to push the benefits of EPON over GPON and vice versa.

Meghan Fuller Hanna is senior editor at Lightwave.

More in Network Design