No Strings For This Kite

How does a small fixed wireless company compete against the likes of Sprint? One startup is looking to beat them in their own backyard using a combination of better technology and service.

by Jim Wagner of internetnews.com [February 28, 2001]

Phoenix is becoming a popular proving ground for fixed wireless providers—consider it the Bonneville Salt Flats of the 2.4 GHz band.

But rather than measuring the speed attained by a piece of sheet metal welded to a roaring rocket, Phoenix is playing host to high-tech companies out to prove the speed and reliability of their wire-free networks. In this speed test, it's data that is sent flying through the airwaves.

Kite Networks is the latest company to erect towers on the hills surrounding the greater Phoenix area. The fixed wireless firm is wrapping up beta testing of its business-class wire-free broadband solution.

Looking to grab a large piece of the Phoenix market, Kite Networks is trying to be all things to all people—offering everything from Webhosting services, streamingmedia feed, video conferencing, Voice over IP, and Virtual Private Networking to businesses—as well as high-speed fixed wireless Internet access to residential users in Phoenix.

Flight of fancy
Kite Networks began testing its network in December 2000, with 30 businesses participating in field trials. Jerry Sullivan, Kite Networks president and chief executive officer, said in each test the company was able to install fixed wireless system in less than the guaranteed five to 10 business days.

As a result of the quick install time and quality of service, most of the beta testing companies signed up for long term contracts to continue connectivity services with Kite Networks.

Salt River Project, one of Arizona's largest utility companies, is still in the process of testing Kite's fixed wireless network. SRP has connected about 100 employees from its IT department to the wire-free broadband service provider.

"The beta test has been progressing successfully and we will continue to look internally for any appropriate application fits within SRP," said David Keough, SRP wireline product manager.

"The installation process was very fast and simple. The Kite network may have the potential to fill any number of rolls where IP centric applications are required for high-speed wireless access and rapid deployment," Keough added.

Kite's wireless solution is proving to be superior to the Multichannel Multipoint Distribution System (MMDS) super-cell technology used by Sprint Corp., a company that's been providing high-speed fixed wireless services for more than a year in Phoenix.

Wireless dogfight
Sprint also tested its first fixed wireless solution in Phoenix in 1999, when it acquired People's Choice TV and its SpeedChoice subsidiary. Sprint has been busy deploying fixed wireless systems in other markets since then, and currently plays host in 13 U.S. Cities.

Last August, while Kite Networks was still called Loop Technologies, the company penned a deal with Cisco Systems, Inc., to become a Cisco Powered Network Program member.

Kite's last-mile solution uses Cisco gear to support data transport over the 5.7 GHz Unlicensed National Information Infrastructure (UNII) band, which doesn't travel as far as Sprint's 2.4 GHz MMDS analog technology.

Because UNII data only travels about 3.5 miles—compared to Sprint's 30-mile service radius—Kite setup micro-cells throughout the city to compensate for the abbreviated transport distance. Jerry Sullivan, Kite Networks president and chief executive officer, said the result of the micro-cell buildout was that Kite gained reliability in data transport.

"Sprint is working on super-cells which is going to make for more interference issues for the consumer, whereas we're running more of a sectorized cell site, which is what you see in the PCS technology," Sullivan said. "It does require more cell sites for our overall infrastructure but will cost approximately what Sprint's MMDS users are going to see."

Float a balloon
Sullivan said he isn't worried about bringing his company a year late to the wireless party started by Sprint, because fixed wireless is just getting started as a viable broadband service.

"I've been in telecom for well over a decade and I've never been first-to-market in any of our cellular one or (personal communications systems) markets, so competition in the area is good," Sullivan said. "We think, given the opportunity to talk to some of our existing customers, even some of which were prior Sprint customers, that we have a network that will meet customer satisfaction."

"In fact, I'd like to see some more competition by Sprint," Sullivan continued, "just to see what kind of growth we could achieve."

According to Allen Nogee, Cahner's Instat senior analyst for wireless component technology, Kite's success is entirely dependent on itself.

"Generally wireless has not been that successful, although there is no reason that can't change," Nogee said. What it all comes down to is service, cost, and reliability."

Cahner's figures that the latency issue, a big problem Sprint and other wireless companies are facing as it deploys, has been resolved.

Box kite
Powering Kite's network are Series 12000 gigabyte Internet Router switches and Cisco 7200/7300 uBR routers and uBR modem cards, a network designed to bypass the line-of-sight requirements of older systems. Data is instead carried via fiber supplied by UUNET.

Sullivan said the configuration gives him a lot of room to work with, unlike a line of sight network, and lets him get away with putting trees, bushes and even plate glass between the two antennas.

"Because we control our own switch there is no latency in the system," Sullivan said. "We're demonstrating a symmetrical range of up to 25 megabits per second.

"Even in a wireless scenario, we are able to provide true broadcast media—three to five channels like the Weather Channel, CNN/fn and Bloomberg—at a very low cost because streaming media actually ties up very little of our network," Sullivan added.

Flight schedule
Phoenix is the first in what Kite officials plan for an eventual nationwide deployment in 2001 and 2002.

Once deployment in Phoenix has been completed, the company plans to start a southern swing through Dallas, Fort Worth and Houston, TX, and Atlanta, GA because deployment costs are generally less expensive south of the Mason-Dixon line.

After that, Sullivan said, Kite Networks plans on delivering its fixed wireless services nationwide.

—End