Get ready for the REAL WiMax, now that Sprint Nextel has finally revealed its plans for deploying the new wireless networking technology in the United States. Having pretty much debunked in earlier columns our expectations for WiMax in general and Clearwire's approach in particular in the U.S., you might be surprised that I see Sprint Nextel succeeding in this same space. What's key (and very different) is HOW they are doing it. It will change the way many of us access the Internet and help make prices lower for the rest.

The problems with WiMax, as I have explained before, are that its higher speeds -- and more significantly its higher speeds over much greater distances -- are mainly dependent on using licensed frequencies that allow higher transmission power. And even for this higher-power licensed service, true data rates drop off significantly with distance from the access point. What we read over and over regarding WiMax is "up to 70 megabits per second at up to 30 miles." This performance, while theoretically possible, will never be met in normal WiMax service and is downright impossible for the unlicensed service that many people envision.

I'm not saying that WiMax is bad, just that it isn't all that it has been promoted to be.

The first national WiMax service in the U.S., Clearwire, suffers from all these problems. Clearwire will find that the $600 million it recently got from Intel and the $300 million from Motorola won't make it into a built out, usable national network. Of course, Clearwire can raise more money or might even finish building the network using operating profits, but that's not the normal model for an industry that can effectively print money, but generally not until a network is running and mature.

So what makes Sprint Nextel different? Everything.

I have long predicted that Sprint Nextel would come to dominate the U.S. WiMax business, primarily because the company has the largest portfolio of licensed frequencies usable for WiMax -- most of those acquired in an early-90's buying spree that cost Sprint more than $1 billion. These frequencies were bought not for WiMax, which hadn't even been thought of, but for Multipoint Multichannel Distribution Service (MMDS), then known as "wireless cable TV." Wireless cable was popular because it was cheap and quick to roll out, but it didn't work very well. Signals were absorbed by water in the air (or in the leaves of trees), so service was spotty when it rained and for homes with trees around them. MMDS also suffered from multipath interference problems that caused ghost images on TV. In short, it didn't work. Sprint recast the service as fixed-wireless Internet, but the same problems remained and eventually that service, too, was shut down with Sprint taking a large loss.

Reason number one, then, why Sprint Nextel's WiMax service is different from its competitors is that the company already owns the frequencies, so the effective cost of acquiring them is nothing. Most competing WiMax operations that intend to use licensed frequencies have to pay for those frequencies, which constitute a large part of their network budget. Meanwhile, Sprint Nextel can just move forward and build its network.

Clearwire's $900 million for expanding its WiMax service nationally beyond its current 30 cities is not only smaller than the $2-3 billion Sprint Nextel plans to spend doing the same job over the next two years, Clearwire has more to accomplish with that smaller pot of money.

A LOT more. Where any new WiMax player would have to build towers or buy space on them and install or purchase network connections to those towers, Sprint Nextel already has a national cellular network in place with thousands of towers in hundreds of cities, all connected through an existing fiber-optic network. Again, they have more money to spend yet less to accomplish.

And here's an important distinction. Sprint's core cellular network is based on Personal Communications System (PCS) technology, which uses less power and therefore shorter ranges than other cellphone systems. This meant that when Sprint built its PCS network in the 1990s, it had to have proportionally more cells with towers closer to end-users. More towers closer to users works well with the weaknesses of WiMax as described at the beginning of this column.

This disadvantage of PCS turns out to be an advantage for deploying WiMax.

Compared to an ab initio national WiMax network, then, Sprint Nextel's job is a lot easier. Rather than building a new network, they can adapt an older one. Rather than buy new frequencies, they can use licenses bought long ago and feared worthless. Even the multipath interference problems of MMDS are handled by the fact that WiMax is specifically designed to work in multipath environments and even to derive greater range from the interference.

Even the way Sprint Nextel announced its WiMax service was different. They positioned it as the company's 4G wireless data service -- a MOBILE WiMax service -- making it a simple upgrade for millions of existing voice and data customers.

It is a lot easier to up-sell an existing customer base (Sprint Nextel already has 51 million customers) than to create a new customer base from scratch.

Why, then, if it appears to have every advantage, does Sprint Nextel still have to spend $2-3 billion? Because for all the company's advantages WiMax is still difficult to do well. Thousands of towers need to be upgraded along with beefing up the supporting network that will have to carry 10 or more times as much data as before.

It's for these reasons that I suspect WiMax in the U.S. will mainly end up being shared between Sprint Nextel and lower-range unlicensed services that may previously have used WiFi. And this is no WiFi killer: WiFi, itself, will be with us for years to come.

This is Sprint Nextel turning lemons into lemonade. It plays to all the company's strengths. Unlike other 3G wireless data providers, Sprint Nextel won't be afraid of people actually using its WiMax system, so they can aggressively market the service. While Sprint Nextel is talking about WiMax as a mobile service, there is no reason why they can't also compete for fixed wireless customers. You don't see Verizon or Cingular, for example, trying to sell their 3G products for use in fixed home or office applications. Some 3G operators, in fact, specifically prohibit using their equipment in that way, because while the network is fast, its total aggregate data capacity is a lot less than you'd expect.

Not so with Sprint Nextel, or at least that's the way it looks from here. With smaller cells and a robust backbone, there is no reason why the company's 4G (WiMax) network can't compete directly with cable modems and DSL, which was the original vision of WiMax , whether it was really practical or not. And as an existing landline voice provider operating now primarily in the backyards of its competitors, Sprint Nextel could mount a devastating VoIP telephony sales campaign using its WiMax network.

All this bodes well for customers, with voice and data prices likely to fall as bandwidth increases. I wish the company well, but, as always, I don't own any of their shares.