Got fiber?

When Dan Akerson took the podium last September to announce that the recently formed company that combined Nextlink Communications and Concentric Network was changing its name to XO Communications, he solidified more than just a change of corporate monikers.

To a large extent, Akerson, chairman and CEO of the new entity, mapped out a plan for a nascent local multipoint distribution service (LMDS) market that had started with a world of promise but had devolved into niche technology.

The announcement, coming less than a year after the merger of Nextlink and Concentric, positioned LMDS not so much as a technology that could leap over tall buildings without costly infrastructure upgrades, but as a single element in an overall access strategy that includes a combination of fiber and wireless platforms.

Indeed, the merger points to a tacit admission by Nextlink that while the company may have held a large swath of spectrum in the U.S., it could not necessarily provide all services to all customers with it. "We had this unique set of assets that we think sets us apart," said Akerson in September. "We had to productize these assets."

And while XO Communications chose to use the occasion to highlight its entry into the gigabit Ethernet market, the implication was the same: Few carriers can survive on wireless alone.

"Long term, we’ve always said that we were going to approach the market with a variety of access technologies," Akerson said. "We’ve seen various applications for different fixed wireless modes of transmission. We’ve used a combination."

Getting in gear

Others have followed suit in combining wireless and fiber infrastructure. In fact, asking carriers such as Winstar, XO and Teligent about blended access strategies sparks discussion about who was first than whether they’re actually implementing the strategy.

In an address at a Salomon Smith Barney conference last month, William Rouhana, chairman and CEO of Winstar, said the company often is lumped unfairly in with wireless carriers, even though its strategy has been to use a variety of access technologies.

"We use fixed wireless, but that’s not all we use," he says. "We are technology agnostic—we always have been—but we’re also people who look at numbers. When we analyze the creation of bandwidth in the local market, we look at

capital. We’re going to use whatever technology will let us create bandwidth that we can own. Every single building is a separate intellectual exercise."

David Ackerman, group executive for network and systems services with Winstar, is more to the point. He says that the company is beyond the debate over technology usage and is more focused on applications and solutions sets for customers. "Everyone is following us. What we really believe is that the network is not the end all itself. We don’t see anyone even attempting to do what we’re doing with applications."

Winstar provides perhaps the greatest example of the evolution of the blended access strategy.

Initially created to bid on the 1300 MHz of spectrum in the LMDS (28 GHz) band, the company spent the first few years after the auction building up its network and focusing on small and medium-sized enterprise users. That has evolved rapidly over the last six months to the point that large enterprise customers are coming to the company looking for physical redundancy in their networks and a quick fix for capacity problems.

Using an arrangement whereby Metromedia Fiber Nework and Williams Communications build the fiber, but Winstar owns and controls it, the company can offer managed services that wouldn’t be feasible over leased facilities.

"[Both Winstar customers] Oracle and Microsoft see the value to their customers of placing their applications on our network," says Ackerman. "It gets them out of selling shrink-wrap products and into selling monthly fees. It would have been foolish for us to think of this two or three years ago, because the network wasn’t ready."

The move to a services-based approach is pure economics, says David Twyver, president and CEO of Ensemble Communications, a manufacturer of LMDS equipment. "You just couldn’t do an LMDS network design that let your business plan close any other way. All these guys were trying to sell T-1s at 30% less than the ILECs. That’s a lose-lose. There’s no elasticity in that market."

Fiber incentives

The blending of wireless and wireline access strategies has become so pervasive that it is seeping into other less traditional wireless approaches that once were viewed as pure niche plays. After launching itself as a pure-play wireless carrier using unlicensed national information infrastructure band frequencies, Fuzion Technologies announced last month that it has signed deals with several carriers—notably AT&T—to provide access to fiber in areas where it is lacking wireless infrastructure.

"This gives us ubiquitous coverage across the U.S.," says Dave Frank, president and chief operating officer of Fuzion. "I’ve got extremely competitive rates in 40 metro areas for wireline and I’ve also got wireless infrastructure being developed. It gives us a total broadband network solution."

The need to sign a fiber deal was almost a given, he adds. Targeting the small and medium-sized business markets, Fuzion often found itself lacking capacity or the plant to get to its customers.

"Fuzion’s target before was forced into the coverage area of the hub," says Frank. "In this arena I have no limitations. The customer doesn’t have to be in the three-mile loop. It put us in a national customer arena, which we couldn’t do before."

Moreover, the company doesn’t have to deal with some of the regulatory issues such as roof rights with every customer. In fact, Frank cites roof rights issues as a major force that is pushing wireless carriers, regardless of spectrum, to acquire fiber. Gaining access to the rooftops, an issue that has been settled largely by the FCC from a regulatory perspective, still is a hindrance at the practical level, says Frank.

"The wireless player needs to obtain building roof rights, and as much as the Telecom Act has attempted to make an even playing field, incumbents continue to have an unfair advantage. We still believe in the wireless play." However, Frank adds, the industry will need help from the FCC to ensure that roof rights issues are created on a more equitable basis.

The blended access strategy also is emerging because LMDS spectrum ended in such few hands, Ensemble’s Twyver says. "I don’t personally believe in the pure play wireless CLEC model. It kind of works on mobile, but you have to look at where the spectrum ended up. The folks just addressing access with radio won’t be successful. It’s a critical technology for extending fiber. It’s just economics."

Depth vs. breadth

Regulatory concerns and the ability to reach out to a larger audience are not the only factors pushing formerly pure wireless carriers to bulk up on fiber. It’s also the desire to offer high-bandwidth services to larger enterprises and re-deploy wireless equipment to less fibered areas. And while vendors are continually improving the capacity of radios in the LMDS band, even the most die-hard wireless bigot isn’t going to ignore fiber if it’s available.

When XO unveiled its new brand, the company timed the announcement to coincide with it launch of gigabit Ethernet services over its existing fiber—something it could theoretically provide over wireless infrastructure.

"Gigabit Ethernet is something a lot of people have been talking about, but if you don’t own the underlying metropolitan fiber assets that we do, this is not an easy service to deliver," says Nancy Gofus, executive vice president of marketing and customer care for XO.

And while it’s possible to offer gig-E services over a wireless platform, the use of fiber is opening up larger corporate accounts, notes Winstar’s Rouhana. One of Winstar’s biggest areas of emphasis has been its ever-expanding network.

At last count, the company had at least 1000 MHz of spectrum in the top 60 U.S. markets along with at least 50 fiber rings and an unspecified amount of local fiber. In total, the company will be in 9700 buildings by the end of the year.

In fact, the company is confident enough in the reach of its network that as of January, sales reps no longer received compensation for traffic that is sold off net. And while some debate the merit of such a plan, it shows a certain confidence in the network.

"We will never be 100% on net. We’ll always be accommodating customers that have nine offices on net and 10 offices off net, and we want to do business with them," says Rouhana. "But the revenue we’re adding is the kind of revenue you want—revenue with margin. For the first time in our lives, we’re using the word ‘efficiency’ in our growth."

An oft-cited figure in the broadband wireless community notes there are somewhere between 750,000 and 800,000 commercial buildings in the U.S. Of those, anywhere from 4% to 7% are wired with fiber. That leaves a minimum of 697,000 buildings with copper or wireless plant. In the international market, particularly Europe, the percentages are lower and are compounded by cultural concerns about digging up historic cities to install fiber.

Getting bigger

"Fiber is growing, but it’s growing mostly on the inner metro market," says Hamid Akhavan, chief technology officer of Teligent, which in May last year signed a deal with Level 3 Communications to provide fiber and network services in some markets. "Nobody’s projecting that fiber penetration in the next decade is going to reach over 7% to 9%. Ninety-five percent of the buildings don’t have access to fiber. If they want multi-megabit service, that’s virtually impossible to get on the copper infrastructure."

Like other fixed wireless carriers, Teligent looks to economics as the primary driver.

"We’ve always believed that radio-based and fiber-based technologies have their place in the market," says Akhavan. "The challenge in the fiber business has always been the number of access points. You may have a fiber snaking through a metropolitan area, but very few buildings are actually on that fiber. The economics of both businesses look significantly better if you can leverage both."

Typically, though not universally, large company corporate headquarters are in fiber-based buildings while field offices rely on copper or wireless infrastructure. In many cases, particularly in third- and fourth-tier markets, the economic justification may never be there to pull fiber all the way to a building. "You have to have a certain amount of telecom spending in the building before you can depreciate the capital needed to lay out fiber," Akhavan notes. The upshot has LMDS carriers winning more large corporate accounts over the last year, sometimes without specifically marketing to them.

"The large business has the same needs; it’s just a bigger decision," says Winstar’s Ackerman.

That situation, however, is changing in part because of capacity improvements in wireless technology. Ceragon Networks, which has developed high-frequency wireless systems, recently introduced a new ultra-capacity system that can transmit at OC-6 (311 Mb/s). Others are looking to unveil OC-12 systems near the end of this.

"When we started delivering product, it was thought in general that bringing eight T-1s would be sufficient to the building," says David Ackerman, president of Ceragon (and no relation to

Winstar’s Ackerman). "The way the architecture would be was high capacity in the backbone and lower capacity in the building. What everybody is realizing is that eight T-1s are not enough. They need to bring a minimum [of] DS-3."

Concurrent with the huge new bandwidth demands of larger enterprises, medium and small-sized businesses are beginning to ask for data pipes that go beyond copper’s capabilities. In introducing its bundled package that includes some very large pipe products, XO says it was aiming at enterprises with anywhere from 10 to 100 lines.

"That’s our sweet spot, that’s our legacy," says Ackerman. "We’re not going to give it up."

Not surprisingly, as carriers’ capacity needs have increased, vendors have followed suit. Alcatel, which is one of the major suppliers of LMDS equipment, is seeing an increased mixture not just of fiber and wireless, but also of different wireless architectures.

"We feel the customers who are successful will be the ones selling a total broadband portfolio," says Dave Kimzey, vice president and general manager of Alcatel’s wireless access products. "The customer doesn’t really care about how it gets there. They want the service."

Specifically, the company sees increased use of point-to-point links that target single buildings or, in some cases, individual companies, says George Hendry, vice president of business development for Alcatel’s fixed wireless group.

"One interesting difference between the U.S. and the rest of the world is [the FCC] gave huge blocks of spectrum. [Domestic operators] can overlay point-to-point links. In Europe it’s more channel pairs. The U.S. spectrum is much more flexible," he says.

Even with that huge chunk of bandwidth in the air, some domestic carriers already are bumping up against spectral efficiency issues.

Teligent, for example, is deploying an even mix of point-to-point and point-to-multipoint radios, says Akhavan.

Initially, the company thought fixed wireless was best suited for more of a broadcast architecture.

"Today we see for the foreseeable future a 50-50 mix, " he says. "We’re going to use all the tools. In the main downtown core areas, we may actually use a point-to-multipoint architecture. Where you have fewer buildings to serve but each one of those has a large customer, it’s more efficient. If you have a large number of smaller buildings, you’d rather share that cost of infrastructure."

Winstar by contrast runs about 80% of its traffic over point-to-point links (or point-to-consecutive-point to get around line-of-sight issues). That will change as bandwidth requirements keep increasing, but, according to Ackerman, the company will likely always have more traffic over point-to-point links. Winstar also is experimenting with some free space optics, though it won’t deploy the technology until it’s more mature, he adds.

The DSL effect

Such technologies will help fixed wireless operators address even the large enterprise accounts. Ironically, small business demand is not abating in part because of the failure of many DSL service providers to deliver on their promise.

"The data side of our business is growing, because the myth of DSL was shattered last year," Rouhana says. When DSL companies approached small businesses promising to deliver T-1 speed for $99 per month, it raised expectations, he adds. Those same providers’ subsequent difficulties gave wireless carriers an opening to provide a proven product, which didn’t have to compete on price.

Teligent has never really considered DSL competition and in fact has adopted wireless DSL as one marketing approach. "DSL was never officially really targeting the same segment we did," says Akhavan. "[DSL providers] were really targeting one notch below where we were."

In essence, Teligent and all other fixed wireless carriers have bigger fish to fry. "We’re deploying as fast as we can. We’re not limited by demand. There are 750,000 buildings out there. We’re cherry picking as we go along. We’re thinking for the next three years or so we’re going to continue to do that. We have an enormous supply of buildings in front of us."