The new CableLabs spec is indeed being called "Docsis 3.1" -- there was some doubt about that -- and it's being publicly discussed for the first time on Oct. 18 at the Society of Cable Telecommunications Engineers (SCTE) Cable-Tec Expo in Orlando. (See CableLabs, SCTE to Shine Light on Docsis 3.1.)
The 90-minute session, aptly titled "Docsis 3.1 Specification and Standard Development," is slated to start at 11 a.m. Eastern time that day, moderated by CableLabs VP of Access Network Technologies Dan Rice. He'll be joined by:
- Jeff Finkelstein, senior director of network architecture, Cox Communications Inc. ;
- Howard Pfeffer, senior vice president of broadband engineering and technology, Time Warner Cable Inc. (NYSE: TWC);
- Jorge Salinger, vice president of Access Architecture, Comcast Corp. (Nasdaq: CMCSA, CMCSK); and
- Matt Schmitt, director of Docsis specifications, CableLabs.
After the roundtable, SCTE Senior Vice President of Engineering and CTO Daniel Howard will come on to explain how the Society will do its part to put Docsis 3.1 on a fast track.
It's expected that the roundtable will cover the technical goals of Docsis 3.1 and will explain how far developed the specs are.
But some details about what the industry has in mind for Docsis's next chapter have been circulating for months.
Kicking out QAM
At The Cable Show in May, engineers from Arris Group Inc. (Nasdaq: ARRS), Cisco Systems Inc. (Nasdaq: CSCO) and Motorola Mobility outlined several ideas on how to refresh Docsis and put it on the road toward capacities of 10Gbit/s in the downstream and 2Gbit/s in the upstream by, in part, expanding the spectrum in both directions. Multiple industry sources say the 3.1 project will look to change the Docsis PHY.
Those discussions appear to be rallying around more efficient, non-QAM modulation technologies, including orthogonal frequency-division multiplexing (OFDM), a scheme that's popular in the wireless world. It relies on small (on the order of 10KHz-wide) subcarriers that should perform better than QAM in noisy conditions, which tend to plague the lower portion of cable's upstream spectrum.
OFDM could be used in the downstream and upstream, but the industry is believed to be chewing on a hybrid approach in which OFDM is applied just to the upstream, while the downstream continues to be based on QAM modulation, so operators can preserve the investments they've made in new edge QAMs. OFDM, by the way, is also being considered for EPON Protocol Over Coax (EPoC), an Institute of Electrical and Electronics Engineers Inc. (IEEE) standard-in-the-making. (See Countdown's on for EPON Protocol Over Coax and EPON-Over-Coax Starts Its Standards Journey .)
FEC steps in
Another technology on the table is low density parity-check (LDPC), a forward error correction (FEC) technique that would also help data transmissions overcome noise, at a bandwidth overhead smaller than that of the current Reed-Solomon approach. The hope is that a combination of OFDM modulation and LDPC could improve spectral efficiency by 24 percent or more.
One possible item of debate is whether Docsis 3.1 will be completely backwards-compatible with Docsis 3.0 (the desired goal), or simply coexist with it. Every previous Docsis spec has been compatible with the versions that preceded it, and that appears to be the direction in which many engineers would like to see for Docsis 3.1. If only coexistence is possible, that could require operators to carve out precious spectrum just for the new Docsis 3.1 platform.
And when will it all come together? "There's a desire to have it happen sooner than reality would let it," says a cable industry source, who doesn't expect initial Docsis 3.1 drafts to appear until sometime next year.
An executive with a vendor familiar with the project agreed that the cable operators are pushing the 3.1 project forward anxiously. "I've never seen the MSOs this united about wanting to do something," he tells Light Reading Cable in a recent interview about the topic.
Until the specs are frozen, expect Docsis 3.1 to morph and change like an amoeba during the next year. But the upcoming roundtable will offer the most solid details on the project yet.
- Will Anyone Switch On Cable's Upstream Booster?
- Metrico Tests Concurrent Voice & Data
- Cable's Upstream Gap
- Cisco Hints at What Comes After Docsis 3.0
— Jeff Baumgartner, Site Editor, Light Reading Cable