November 2004

Why WiMax?

Continued from page 1

By Wade Roush

It was Intel's announcement of a major push into WiMax technology in January 2004 that helped the standard emerge into the spotlight. The company's Centrino initiative had already put Wi-Fi chips into millions of laptops. "After we did that, we began looking at whether you can cover full cities with Wi-Fi," explains Scott Richardson, manager of Intel's broadband wireless group. While it would be relatively simple to blanket an entire city with Wi-Fi hot spots, the company decided, such a patchwork would be difficult to administer and would operate over too narrow a frequency range to deliver sufficient amounts of data for future needs. "We came to the conclusion that Wi-Fi needed to evolve into more of a 'carrier' technology, deployed by a service provider, and needed to exploit a lot more spectrum options," says Richardson. WiMax, which operates at greater distances and over a greater range of frequencies, turned out to be ideal.

The company began designing communications processors to exploit these frequencies -- from roughly two to 11 gigahertz, a range used mainly by Wi-Fi, microwave ovens, and certain types of radar -- and had delivered the first sample chips to manufacturers by September. Meanwhile, it began promoting an industry association called the WiMax Forum to certify equipment from vendors as WiMax compliant. And through Intel Capital, the company's venture wing, it has begun to make strategic investments in a few companies that plan to demonstrate how WiMax can be put to profitable use.

Seattle-based Speakeasy is one of those companies -- and a prime example of the economics driving WiMax's rollout. Founded in 1994 as an Internet café, Speakeasy has evolved into the leading provider of extrafast DSL connections to hard-core online gamers and technical professionals who work at home. But because DSL works over phone lines, it has an inherent limitation: about 30 percent of residences in the cities Speakeasy serves are too far away from phone network central offices to get a usable signal. "That's a lot of people to turn away," says Speakeasy CEO Bruce Chatterley. "That's why we started to look for alternatives, and that's why WiMax is so strategic to our business." Speakeasy will begin technical trials of WiMax equipment using Intel's chips by the end of this year and hopes to offer broadband wireless connections to business and residential customers by the middle of 2005.

But while the emergence of WiMax will give consumers, businesses, and people in hard-to-reach areas a powerful new way to connect to the Internet, it won't happen overnight. For one thing, it could take manufacturers some time to reach the economies of scale that would enable consumer-priced WiMax equipment. Then there's the cost of building a network of transmitters. "People tend to think that you can put one WiMax tower on a hillside and beam around the entire city, and that's certainly not the case," says Intel's Richardson. "When you fill up a cell, you use up the capacity" -- meaning that providers will still have to add towers as demand grows, just as they do in traditional cell-phone networks.

But TowerStream, a Waltham, MA, company that plans to add WiMax to its existing broadband wireless services, thinks it has that problem licked: it's already tied up what chief operating officer Jeff Thompson calls "beachfront property" atop many of the tallest buildings in Boston, New York City, Chicago, and other cities, and it will simply install the new WiMax-certified gear alongside its existing transmitters. "When WiMax comes out," Thompson says, "our speed of deployment will be very quick. We'll have a wireless backbone in the sky." Which sounds great -- as long as it really does cost less to use than our earthbound skein of wires, fibers, and cables.