Multichannel Multipoint Distribution is at a critical juncture.
Industry leaders Sprint and WorldCom have found that using first-generation Multichannel Multipoint Distribution System (MMDS) gear to deliver high-speed Internet access doesnt deliver wide enough profit margins. So theyre pushing vendors for second-generation near- and non-line-of-sight technology.
But its not a sure bet that even this cutting-edge technology will enable them to make a profit.
So far, Sprint and WorldCom have used first-generation "big stick" equipment — often a single tall tower — to cover an entire town. But the big stick has its shortcomings: Its a line-of-sight technology, which means in many markets only half the population can be served. And to serve those customers, technicians must install rooftop antennas, which is an expensive undertaking.
Vendors have begun to develop near-line-of-sight and non-line-of-sight technology that will give operators greater reach, but the migration to second-generation networks isnt well-mapped. While near-line-of-sight was designed to serve as an interim solution, it may be skipped.
"The fact that they havent deployed any near-line-of-sight is telling us a lot," says Andy Fuertes, an Allied Business Intelligence analyst.
Hybrid Networks and Vyyo have developed near-line-of-sight gear, but their products havent hit the market yet.
The difference in coverage between near- and non-line-of-sight is minimal — 5 percent to 10 percent more of the population can be reached with non-line-of-sight, Fuertes says. And since non-line-of-sight gear is likely to become available very shortly after the near-line-of-sight products, operators may just wait for the better coverage. In addition, some non-line-of-sight products can be self-installed by users, eliminating a major cost for operators.
"We havent seen anything thats near-line-of-sight whose path to non-line-of-sight is certain enough for us to make the jump," says Russ Wiseman, senior vice president of Internet operations of MMDS operator Nucentrix Broadband Networks.
Put the Stick Down
Earlier this year Sprint said it would stop building out new markets using the big-stick architecture while it waits for the new gear to be ready, and has promised its first second-generation markets will be functioning by the end of 2002. While a number of vendors say their next-gen equipment will be commercially available by the end of this year, some operators dont expect them to meet that deadline.
But even after operators have rolled out the new technology, its not clear it will pay off. "Once we have non-line-of-sight, our problems dont go away," Wiseman says. "Fundamentally its a question of how much investment we need in a market to achieve coverage." Non-line-of-sight technology requires a cellular architecture, similar to that used by mobile voice operators, and in most markets this means operators should be able to attract enough customers to build a profitable business.
Vyyo says near-line-of-sight technology, which requires fewer cell sites, is suitable for operators that, like WorldCom, are targeting small-business customers.
"WorldCom is going for the business market, so they dont need small cells because the number of users theyre going after is more limited," says Arnon Kohavi, senior vice president of strategic relations of Vyyo. WorldCom has used Vyyo gear in cities that include Memphis and Chattanooga, Tenn., and Bakersfield, Calif.
Operators would be smart to chase the business market, Kohavi says. "To be honest, Im not sure any carrier has the financial resources to commit the money up-front to build a network for residential users."
Hybrid Networks, which supplies the majority of the big-stick deployments today, has also developed a near-line-of-sight technology that will be commercially available in the fourth quarter, says Malik Audeh, Hybrids director of wireless systems.
Audeh is betting on big-stick architecture even when near- and non-line-of-sight products become more widely available and proven. In a city like Chicago, where antennas can sit as high as the Sears Tower, operators can start offering service quickly to lots of customers without a massive infrastructure investment, he says.
IPWireless has developed non-line-of-sight equipment for the residential market based on the universal mobile telephone system standard defined for mobile third-generation networks. "Components designed and produced in mass market volumes for UMTS can be used in our technology," says Roger Quayle, IPWireless chief technical officer.
The IPWireless solution includes a battery-powered modem that customers could use for access anywhere in their homes or neighborhoods.
MMDS gearmaker NextNet Wireless says it has an edge because it started out developing non-line-of-sight equipment. "Were the only ones that have matured to the point where we can build a cost-effective unit based on [Application-Specific Integrated Circuit] technology," says Chuck Riggle, the companys vice president of marketing and business development.
Other vendors are still building products based on field programmable gate array chips, which allow manufacturers more flexibility in development, but are pricey, he says.
Operators choosing between near- and non-line-of-sight gear have found it difficult to determine if vendor gear meets their qualifications and operates well in the geography of different markets. But recently, the Institute of Electrical and Electronics Engineers built a number of channel models against which operators can judge equipment.
"For the first time with the IEEE channel models we have a yardstick that service providers can use to quickly determine whether a vendors product is truly non-line-of-sight," says Bob Powell, director of product management of Iospan Wireless.