Hollow Fiber from OmniGuide
I believe that we've discussed developments in hollow fiber [not blown fiber tubing (we've discussed that, too), but fiber that is itself hollow] from OmniGuide, a spinoff company from MIT. Here's another article about same, courtesy of VidiVice on the HLIT Thread:
business2.com
Interestingly, the hollow permits an improved index of refraction. How, then, does this alter the answer to my riddle in LMT? See message 9982 in LMT and the replies that follow.
====================
The Hollow Fiber article from Business 2.0 Mag, for posterity:
"Super Cables"
Researchers at MIT have found a way to make fiber optic cables thousands of times more efficient–and a lot cheaper to deploy.
From the December 26, 2000 issue, by Robert Poe
--------
NOTE: View the "Optical Cables Go Hollow" Xplanation is at: business2.com , complete with graphics.
--------
Progress in fiber optic technology–the Internet's plumbing–is advancing at breakneck speed. The number of data bits being stuffed through today's hair-thin glass fibers is doubling every 6 to 12 months, and ever-faster switches are being developed to route data streams. Still, high-bandwidth applications, especially those involving video, are already threatening to devour every gigabit of capacity. And the broadband age has barely begun.
However, a breakthrough from the Massachusetts Institute of Technology may produce a dramatic increase in the capacity of fiber cables–and bring the goal of unlimited bandwidth a giant step closer.
In essence, researchers have found a way to replace solid glass fibers with hollow fibers containing mirrored inner walls. The new design could multiply cable capacity by thousands of times, and cut the costs of long-haul networks by millions of dollars. "If it's as good as they say it is, this is a revolutionary, not evolutionary, technology," says Mark Langley, director and senior research analyst of communications equipment subsystems for investment bank Epoch Partners in New York.
Mirror, mirror in the ground
The MIT researchers have developed mirrors that reflect virtually 100 percent of light striking them from any direction, compared to less than 95 percent reflected by conventional mirrors. Theoretically, at least, hollow tubes outfitted with mirrored walls could guide laser-generated light pulses carrying encoded data thousands of miles without expensive amplification equipment.
Although the lab work is promising, researchers have yet to prove that such cables can be manufactured commercially. OmniGuide, a company founded by the MIT researchers to manufacture the technology, has yet to build a working prototype.
"We believe we can manufacture this fiber," says Uri Kolodny, co-founder and director of marketing. "Obviously that is something we still need to prove to the world."
Toward that end, OmniGuide has gathered an impressive team of backers: Ray Stata, co-founder of semiconductor manufacturer Analog Devices, and Mukesh Chatter, founder of terabit switch manufacturer Nexabit Networks. Together the two men have invested a total of $4 million. Kolodny says the next round of financing will be used to set up a manufacturing facility.
No more dirty signals
Although today's fiber optic cables are made of pure, solid, ultra-thin glass, they distort light as it moves toward its destination. This distortion causes different wavelengths, or colors, to interfere with one another, thus limiting the number of wavelengths that can be transmitted at once.
Since the most common method of increasing cable capacity is to send more wavelengths through each fiber–each wavelength carrying 2.5 to 10 or more gigabits of data per second–attenuation continues to plague transmission. Kolodny claims that "several orders of magnitude"–that is, thousands–more wavelengths could be sent through a hollow cable than through glass fibers. The most advanced glass fiber systems currently under development transmit from 80 to 160 wavelengths per fiber; at 10Gbps per wavelength, that means a single hollow fiber could carry as much as 800,000 to 1,600,000 Gbps, compared to a "mere" 800 to 1600 Gbps transmitted by glass systems.
The problems caused by attenuation don't end with a ceiling on capacity: Attenuation also requires manufacturers to install equipment to receive the distorted or "dirty" signals and send them out "clean." These signal regenerators can cost tens of thousands of dollars to install on land; underwater regenerators can cost hundreds of thousands of dollars apiece.
What's more, glass fiber absorbs some of the light passing through it, making it necessary to amplify optical signals every 50 to 75 miles or so along the route. (Boosting the strength of the light so it can travel farther without amplification increases interference-causing distortion). Average costs of optical terrestrial amplifiers range from $25,000 for short-haul systems to $53,000 for long-distance models, according to research firm RHK of South San Francisco, Calif.
Hollow fibers, however, could mean the end of such costs. "We are talking about the elimination of both regeneration and optical amplification," Kolodny says. The savings would be massive. For a 32-fiber cable spanning 2,500 miles, eliminating optical amplifiers alone could save from $56 million to $85 million.
As far out as these claims may sound, OmniGuide's founders say that hollow-fiber technology could be put to use in the next few years. The company expects to build a prototype cable within a year, with commercial production to start shortly thereafter. Kolodny also claims that manufacturing costs will be comparable to those for conventional cables.
Epoch's Langley is one of the few analysts who have begun to look seriously at this technology, but he is reluctant to handicap OmniGuide's chances for success. Yet, over the last six months or so there has been a noticeable increase in the number of scientific papers on the technology, "a leading indicator," Langley says, "of what's on its way to market."
Robert Poe (rpoe@business2.com) is a Senior Writer for Business 2.0. |
