x
Optical/IP

AlcaLu Pushes the Optical Envelope

Alcatel-Lucent (NYSE: ALU) flexed its optical muscles this week, announcing a world-record optical transmission and demonstrating other optical milestones in a series of post-deadline papers presented at OFC/NFOEC. (See AlcaLu Marks Optical Record and AlcaLu Pushes 100 GigE.)

In one post-deadline paper released today, AlcaLu announced that it transmitted 25.6 Tbit/s of optical data over a single fiber strand using 160 Wavelength Division Multiplexing (WDM) channels. The transmission went through three 80km spans, using WDM in both the C and L wavelength bands.

Researchers used polarization multiplexing in each wavelength to double capacity, and Raman amplification was used to increase the optical signal-to-noise ratio.

AlcaLu also sought to exploit spectral efficiency, which it did by using an advanced signaling format called RZ-DQPSK (return to zero differential quadrature phase-shift keyed). This allowed it to realize 3.2 bits/second/Hertz (b/s/Hz) of spectral efficiency, compared with commercial systems that generally operate at spectral efficiencies between 0.2 and 0.4 b/s/Hz.

Alcatel-Lucent took advantage of research expertise from both sides of the organization, as researchers from Alcatel's Research and Innovation center in France paired up with a team from Lucent's U.S.-based Bell Labs to work on the project.

The previous record -- a 14-Tbit/s transmission -- was set in September 2006 by NTT Group (NYSE: NTT). (See NTT Demos 14 Tbit/s and NTT Plans 10-Tbit/s Network.)

In addition to its record-breaking optical transmission news, the company touted other work aimed at pushing the limits of existing optical networks.

For example, Bell Labs demonstrated the ability to upgrade 40-Gbit/s long-haul networks to 100 Gbit/s. Using WDM channels typical of 40-Gbit/s networks, researchers transmitted 10 100-Gbit/s WDM data channels over a 1,200km optical transmission distance, achieving spectral capacity of 1 b/s/Hz and effectively doubling throughput.

Martin Zirngibl, director of Bell Labs Data/Optical Networks Research Department, said his team used DQPSK to achieve this. DQPSK, he says, "has much lower bandwidth requirements… about the same as 40 Gbit/s," which could ultimately allow carriers to reuse existing 40-Gbit/s components for 100-Gbit/s applications.

Zirngibl believes interest in 100-Gbit/s will be driven by increased adoption of video on demand (VOD) services. "If you look at VOD -- if it takes off, carriers will need to upgrade their networks. In a few years, 90 percent of traffic will be VOD, and bandwidth demand will go up by a factor of 10."

Heavy Reading analyst Sterling Perrin agrees, saying that there's heavy carrier interest in the technology. "In a survey of 72 operators published in February, 45 percent of operators surveyed said that 100-Gbit/s is 'under investigation' right now and only 13 percent expressed 'no interest' in the technology," Perrin says.

— Ryan Lawler, Reporter, Light Reading

optodoofus 12/5/2012 | 3:11:08 PM
re: AlcaLu Pushes the Optical Envelope > In a few years, 90 percent of traffic will be
> VOD, and bandwidth demand will go up by a factor
> of 10.

This is probably one of the dumbest quotes I have ever seen. Certainly, VOD is popular today and will become more popular as time goes by, but exactly why would anyone be pushing all this VOD traffic across the optical backbone? Maybe if I put all my VOD servers in one place and fed the entire country from that location, VOD could have that kind of impact. Maybe. But that would be a ludicrous architecture. VOD content is already housed on distributed servers today, and will become more de-centralized as it scales. Anyone who believes that VOD will drive this kind of optical backbone demand is either hitting the ganja pretty hard or seriously confused about how VOD works. Or both.

optodoofus
Ryan Lawler 12/5/2012 | 3:11:08 PM
re: AlcaLu Pushes the Optical Envelope To be fair to Martin, he wasn't referring to the type of VOD services that are available today, whereby a limited selection of mostly standard-definition programs are stored and cached locally. He was talking about the possibility for extremely large libraries of high-end video available over the Net. He compared the services driving this demand to Netflix -- which has some 70,000 titles -- and acknowledged that while the most popular content will still be stored closer to the premises, the other tens of thousands of titles probably won't be.
Ryan Lawler 12/5/2012 | 3:11:08 PM
re: AlcaLu Pushes the Optical Envelope To be fair to Martin, he wasn't referring to the type of VOD services that are available today, whereby a limited selection of mostly standard-definition programs are stored and cached locally. He was talking about the possibility for extremely large libraries of high-end video available over the Net. He compared the services driving this demand to Netflix -- which has some 70,000 titles -- and acknowledged that while the most popular content will still be stored closer to the premises, the other tens of thousands of titles probably won't be.
Belzebutt 12/5/2012 | 3:11:07 PM
re: AlcaLu Pushes the Optical Envelope In a few years, 90 percent of traffic will be VOD


When he means VOD, I assume he includes all the pirated movies on bittorrent and usenet that make up half of the internet traffic today.
Sterling Perrin 12/5/2012 | 3:11:07 PM
re: AlcaLu Pushes the Optical Envelope We've surveyed operators separately about their regionaal/metro networks and their long haul/backbone networks. What we've found is that the #1 capacity driver in the regional/metro is video. In the backbone, Internet traffic was #1, and video did not factor in that highly at all. However, as the poster points out, there is a lot of Internet video (YouTube, etc.) in the Internet mix that is hard to single out.

Ultimately, video drives capacity expansion in all parts of the network, but it's a question of closed network video vs. Internet video.

Sterling
boozon 12/5/2012 | 3:11:04 PM
re: AlcaLu Pushes the Optical Envelope It's nice to see a genuine hero experiment, which has the sweet taste of the bubble era.
Let's hope that it'll have more market impact than the previous ones!
HOME
SIGN IN
SEARCH
CLOSE
MORE
CLOSE