COVER STORY- The next killer access technology: LASER Is it reliable? Is it safe?
americasnetwork.com
Laser ? or fiberless optics ? is being touted as the biggest new technology in two decades. Proponents say the technology could put LMDS and fiber tails out of business. But many telecom operators remain unmoved.
By Grahame Lynch
Imagine the following. A technology that allowed you to provide broadband without digging trenches or buying spectrum. A technology that you could install in minutes on a rooftop or in a window without any need for FCC or municipal approvals. A technology that allowed you to never buy a leased line or depend on another carrier?s tail ever again. A technology where the cost of a new customer installation could be as low as $150.
Laser developers say they?ve solved reliability, safety issues
Involvement of Nortel, Lucent lends credibility.
TeraBeam says laser is a CLEC buster, but doubts remain.
By now, most telecom executives are very interested. But then they hear the solution. Laser. And they laugh. Yeah, sure. But what about the fried seagulls? The blinded pedestrians? What about when it rains? Or when the wind blows?
Laser has a big credibility problem as an access technology. But a surprising number of savvy technology companies have been working on optimizing laser as a commercial communications alternative, leveraging off research conducted over decades by the US Military, NASA and foreign governments. Now their work is being picked up by supplier giants such as Lucent and Nortel. And a couple of competitive local exchange carriers (CLECs) are planning to use laser as their secret weapon against the fiber and local multipoint distribution system (LMDS) barons.
FIBERLESS OPTICS
It?s precisely because of laser?s negative connotations that its proponents elect to promote it using a wide range of euphemisms ? "invisible fiber," "free-space optics" and "infrared broadband." Collectively, they are referring to a technology that is little more than fiber optics without the glass.
The major difference comes from the replacement of fiber by atmosphere as the conducting medium. This reduces potential link length from thousands of miles to under 10 miles as a result of factors such as the curvature of the earth, building movement, light-blocking atmospheric conditions and safety restrictions on laser intensity.
But the similarities between fiber and laser are more meaningful. Laser can provide capacities of hundreds of megabits and potentially, gigabits per second ? much superior to the capacities available over MDS and other microwave technologies.
---------
The problem of shrinking and expanding buildings has been completely solved. Rain is not an issue, and neither is snow.
--Dan Hesse
---------
Laser is not new. It previously wasn?t considered a viable technology as a result of the overblown claims made for it, according to Jim Dunn, CEO of San Diego laser manufacturer AirFiber.
"In the past, all the major carriers have tried laser. But they saw it as bad news. Vendors made exaggerated claims about the distance possible on laser links. They talked about 3-kilometer [spans] and 99.999% availability, which was just not possible," he says.
"These equipment suppliers chose not to deal with the problem of building movement," Dunn adds. "The fact is that buildings are affected by heat, especially when the sun is shining on one side. And buildings move in the wind."
The solution is to mount the laser equipment on a steerable electronic mount that is regulated by a tracking database and measuring system. Dunn boasts that his company?s system is so sensitive that it can track a building in a 100-mile-per-hour wind without any signal loss.
A similar steering system is employed by Seattle CLEC and laser manufacturer TeraBeam. Says CEO Dan Hesse, "the problem of shrinking and expanding buildings has been completely solved."
NEVER MIND THE RAIN
So what of the other reservations about the effects of weather on laser transmissions? Rain, for one, provides misconceptions, says Hesse. "Rain is not an issue, and neither is snow, unless you have a complete whiteout."
Dunn agrees: "Rain is not a heavy attenuator of light. Rain droplets have a millimeter wavelength, so they do affect MDS transmissions."
In one test, AirFiber shot laser through a very heavy rainstorm and found that the maximum signal loss was 45 dB/km. Dunn says that transmissions would not be completely blocked until such loss reached 200 dB/km.
Another test, conducted by Washington?s Infrared Communication Systems in an unspecified Southwestern US city, managed 99.9879% reliability across a month, including five days of heavy rain and two of thunderstorms. The major cause of loss? A dirty lens, caused by the proximity of one of two links to a building exhaust fan.
Dunn says AirFiber has optimized its laser units to deal with such instances. They incorporate both heating and cooling units to deal with condensation, as well as solar shields to protect the lens from sunlight and, at the suggestion of investor Nortel, Gortex-patched holes that regulate consistent humidity levels on the inside of the unit.
FOGGY BOTTOM
But fog, which does block light, is a problem for laser. "Fog is like a brickwall," concedes Dunn. But even then, all it does is diffuse light, allowing short links to remain up and running. "There is no major impact from fog for links under 200 meters on our technology."
TeraBeam?s Hesse agrees that fog can block transmissions. It has deployed its first network in Seattle, one of the foggiest cities in the US, in order to test worst-case scenarios. TeraBeam provided the connectivity for last year?s Internet 2 conference in downtown Seattle, and according to Hesse, "not one packet was dropped."
Dunn says that optimal link lengths tend below 200 meters for extremely foggy cities such as Fresno, Calif. and Seattle. For cities with less fog, such as Rio de Janeiro and Tokyo, optimal link lengths range between 350 meters and 450 meters. Hesse is more optimistic about link lengths: "It can go for five kilometers in a nice climate," he claims.
Planning for fog can be aided by accurate weather information, according to Hesse. To that end, TeraBeam is compiling databases based on airport weather statistics dating back 50 years. Marrying these data with a 3D network design tool developed by TeraBeam helps network managers to anticipate and prepare for any extreme weather impacts, says Hesse.
NO FRIED CHICKEN
Another concern about laser is its potentially damaging effects on human eyesight and passing birds. This is a complete nonissue, says Dunn. "Laser is subject to Food & Drug Administration licensing requirements and we?re well within the maximum permissible limits," he says. The laser light used in AirFiber and other related technologies is infrared and invisible to the naked eye. "It has less power than a laser pointer and it definitely can?t hurt birds," Dunn says.
A related problem is the potential outages that can result from passing birds in flight. This is also a nonissue, says Dunn. "The maximum outage is a few milliseconds and on a packet network, that just causes a slowdown."
TeraBeam deployed its first network in Seattle, one of the foggiest cities in the US, in order to test worst-case scenarios. Its 14-link network now covers 70% of the downtown core.
Hesse concurs. In a demonstration to America?s Network of HDTV transmission over a laser-enabled packet connection, an 80% blockage of the receiving unit?s lens merely caused the refresh rate of the image to slow. "The beam is so wide that it?s unlikely a bird could completely block it." Hesse claims that TeraBeam?s lasers are so capable that they will even operate through tinted glass!
TECHNOLOGY 1, MARKETING 0
So if laser is so technically robust, why does it have such a modest adoption rate? It?s partly to do with the tiny size of most of the laser developers. The various manufacturers across the country are little more than engineering startups, with tiny sales and marketing arms. But despite their low profile, they have been surprisingly successful at selling installations to campus enterprise networks. Among notable users of laser technology are Seattle dot-com company Avenue A, New York?s Rikers Island Prison and Hawaii?s Servco Pacific. Infrared Communications Systems even claims to have installed a laser link for Boeing?s Kansas plant that crosses a flight path!
The small laser developers are now attracting interest from the big end of the telco vendor town. Nortel was sufficiently impressed by AirFiber?s technology that it took a lead role in a $37.5 million equity funding syndicate that also boasts Qualcomm as a member. "Nortel lends us a cachet of credibility," admits Dunn.
Meanwhile, Lucent has agreed to manufacture and badge TeraBeam?s laser equipment under a $450 million agreement that will see Lucent contribute cash, engineering staff and intellectual property. TeraBeam intends to keep its technology for its own use at first, but says Hesse, "anything?s possible" for the future.
MORE ACCEPTANCE ABROAD
To date, American carrier interest in laser has been low. America?s Network can only identify one major US operator that uses the technology: Sprint PCS. But acceptance has been much greater overseas, especially in Europe, where cellular carriers use laser as a cheaper alternative to leased line tails and a more reliable alternative to microwave.
Among major cellular carriers using laser are GSM Plus in Belgium, Egypt?s Mobinil and South Africa?s Vodacom. They have all sourced their equipment from England?s PAV Data Systems.
A Gibraltar ISP uses Jolt?s laser technology to avoid high leased line charges.
Another significant international operator using laser is Davnet, which runs CLEC operations in Australia, Hong Kong, Singapore, Chicago, Manhattan and San Francisco. Davnet uses laser to service about 20% of its Australian connections. CEO Stephen Moignard says the platform compares well with its other deployed technologies, which include microwave, fiber, high-bit-rate digital subscriber line (HDSL) and integrated services digital network (ISDN). It?s likely that Davnet will deploy laser in its US networks.
Davnet sources its laser technology from Israel?s Jolt Technologies under an exclusive licensing agreement. Jolt has also supplied other installations in the US, Japan and a particularly interesting customer in the English colony of Gibraltar, at the bottom of Spain. The customer, an Internet service provider (ISP), uses laser to avoid high Gibraltar leased line charges by connecting across the border to a cheaper Spanish alternative.
TERABEAM?S BIG PLANS
Probably the most ambitious plans for laser in the United States come from TeraBeam. CEO Hesse was a high-profile convert from AT&T Wireless, where he was also CEO and, infamously, walked away from $50 million in stock options just weeks before its recent float. The subsequent deal with Lucent has netted TeraBeam some $500 million in cash. With its laser offering, the company plans to take on MDS carriers and CLECs in some 100 US and international markets by 2004.
For Hesse, the selling points of laser are based on pure economics. "We can go to any building in two or three days. Maybe LMDS can do that, but they have the cost of spectrum plus they need access to the riser. We don?t have to have rooftop rights or rights-of-way because we can go through the window."
"We also have a capacity advantage. How many OC-12s are provided to buildings today? We can go at much higher speeds than the alternatives."
Hesse says his background in cellular informs his view on the technology?s potential. "The reason the cellular roaming market was small was because availability and competitive pricing were not there.
"We estimate that less than 2% of the United States? 750,000 office buildings are ?lit? by fiber or LMDS. With our offering, there is no idea what the market might really be like for high capacity."
LASER?S COST ADVANTAGE
There?s no doubt that TeraBeam will be able to extend its footprint for much cheaper cost than the LMDS and fiber carriers.
According to modeling prepared by AirFiber, a laser-based optical mesh network covering some 200 buildings in Boston would cost about $20,000 per building. The mesh network would ensure full redundancy, with an average link length of 55 meters and a maximum link length of 200 meters. By comparison, a fiber network would cost between $50,000 and $200,000 per building.
List prices from rival laser manufacturers fall in a similar $15,000 to $20,000 price range. Although TeraBeam?s costs are still highly theoretical and dependent on scale, it currently runs a 14-link mesh network in Seattle that covers about 70% of the downtown core. Based on industry price standards, such a network could probably be commercially created for under $300,000. A fiber network covering the same area could cost as much as $3 million.
Laser?s cost advantage against MDS providers comes from spectrum savings. Because laser operates in the terahertz range, it is not subject to FCC licensing. Hesse says that spectrum is unlicensed above the 600-GHz range because it cannot interfere with other transmissions. He concedes that MDS carriers may place pressure on the FCC to regulate laser, but all the legal precedents point the other way. "I have a big cost advantage because spectrum is expensive and I don?t have to pay it off."
NO CAPITAL OVERHANG
Laser enjoys yet another cost advantage over fiber and MDS, says Hesse. "We only deploy capital on a success basis when we have a customer. We don?t have to put fiber or cells in first. I have a much closer alignment between my capital spend and my cashflows than my competitors."
"We know how much it costs to produce the equipment and rent the real estate. Our only real cost variable is our number of customers," Hesse adds.
AirFiber is targeting a similar sweet spot in the market. "Competitive carriers are so constrained by budgets that they have to find ways to build networks quickly," Dunn says.
Laser?s cost advantage against MDS providers comes from spectrum savings. Because laser operates in the terahertz range, it is not subject to FCC licensing.
"Fiber is expensive and slow. Not only does it cost $150,000 per building but it can take between a few months and a year to get all the permits and dig up the street," Dunn explains.
For this reason, Dunn thinks laser will be especially popular in Europe and Asia, where city authorities are much less likely to allow competitive carriers to dig up congested streets. AirFiber is moving to quickly globalize its offering by conducting network trials in Madrid, Brussels and Tokyo. Its technology is also being trialed in Dallas, Chicago, Denver, Washington, D.C. and Portland.
======================================================
SIDEBAR:
Laser?s promoters and doubters
Laser sounds too good to be true. For some, that?s why they dismiss it. One recent trade magazine column described the technology as "hot air," adding that it?s been around ? and unsuccessful ? for at least 13 years. Another writer for Forbes magazine went further, suggesting that even some of the current claims for the technology, such as TeraBeam?s claimed point-to-multipoint capacity, are untested.
But laser has also attracted its share of high-profile boosters. Foremost is George "Telecosm" Gilder, who devoted much of the March edition of his technology newsletter to a glowing endorsement of TeraBeam. He describes laser as the "most disruptive and most redemptive technology in all communications." Another is self-styled "prosultant" David Isenberg, who has even helped TeraBeam out with a contribution to their corporate video ? "there is no financial relationship," assures TeraBeam CEO Dan Hesse.
Laser has also attracted endorsements from people not renown for their enthusiastic search for the Next Big Thing. Salomon Smith Barney analyst Jack Grubman has said positive things about the technology, while AirFiber touts an analyst reference list that boasts the likes of the Yankee Group.
But the real question marks about laser focus not so much on the technology, but the people pushing it. Both Forbes and the Economist have reported at length on the colorful background of TeraBeam?s Founder and Chief Technology Officer Greg Amadon. The Economist describes Amadon as a "serial inventor and one-time photographer." Forbes says Amadon lost "lots of money" in a virtual reality goggle venture, in between renting cellular phones at an airport and working for CBS as a cameraman.
The defection of Hesse from the CEO position of AT&T Wireless to TeraBeam has also raised eyebrows. Hesse denies that he left because he was passed over for the spin-off unit?s CEO position in favor of John Zeglis, but because he was sold on the concept by Amadon. "I never knew laser had come this far," Hesse says. Hesse has already paid for himself ? he wrangled Lucent?s cash investment through his personal contacts. He headed Lucent in Benelux Europe for several years.
There are also doubts over TeraBeam?s business model, which combines technology development and service provision in the one company. Isn?t that the discredited business model that caused Lucent and AT&T to split?
Hesse says in response, "If this technology is truly disruptive, why wouldn?t you do both? You can be the only game in town."
That said, Hesse concedes the enormity of the company?s challenge. "We?re building a big team that?s covering manufacturing, engineering, design, real estate, information systems ? I don?t underestimate the job of going from scratch to a global telecom player."
"There?s a cookbook for a cellular startup, but there?s no learning curve for us to take advantage of," Hesse adds.
====================================================
SIDEBAR:
A flock of suppliers
Although TeraBeam and AirFiber have attracted the bulk of publicity about laser after receiving high-profile investments from Lucent and Nortel, they are far from the only game in town.
Among other significant suppliers are San Diego?s AstroTerra Corp., which had its own tie-up with Lucent before the TeraBeam announcement. AstroTerra?s laser gear works at speeds of up to 622 Mbps and one of its reference customers is no less than Fore Systems. Among AstroTerra?s current R&D projects are a 2.5-Gbps link that can reportedly operate error-free at distances of 2.4 km and a platform that would enable ground-to-satellite communications via laser.
Another interesting player is Pennsylvania?s LSA Photonics, which claims a product that can operate at 155 Mbps at distances of up to 14 km. Colorado?s Lightpointe Communications is more of a technology agnostic, offering a hybrid system that combines both infrared and microwave platforms.
Meanwhile, Washington?s Infrared Communication Systems claims 4,000 installations. That deployment number makes the company much more successful than its competitors, which seem to boast of only a few customers each.
|
