eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
On a blustery day in May, Intel scattered employees across Las Vegas to prove the merits of the wireless broadband technology known as WiMax.
The far-flung demonstration included live audio and video feeds from a golfer at the edge of the city, an Elvis impersonator cruising the Vegas Strip, and a guy perched atop the 1,149-foot Stratosphere hotel—where Intel Corp. had placed four directional antennas that made the demonstration possible.
At the controls was Sean Maloney, Intels Mobility Group general manager. From a conference room stage in the Mandalay Bay hotel, Maloney communicated with co-workers, showing a crowd of Interop trade show attendees that WiMax, while a bit choppy at times, is the real deal. All in all, it was a successful demonstration—but it wasnt all WiMax.
While the demonstration boards in the Mandalay Bay used an Intel chip set based on the IEEE 802.16-2004 standard, the bulk of the equipment in Intels demonstration comprised prestandard base stations from Alvarion Ltd. The demonstration itself ran in an unlicensed frequency band.
Intels CEO says WiMax is competitive with DSL and cable.
Click here
to read more.
“Pre-WiMax doesnt mean anything,” said Mark Whitton, general manager of wireless solutions at Nortel Networks Ltd., in Brampton, Ontario, which is a proponent of WiMax and a member of the WiMax Forum. “It means its not WiMax. Were still a ways away.”
For the past two years, the nascent WiMax technology has been something of a broadband media darling, promising versions that would offer both a last-mile substitution for a land-line Internet connection and a muscled-up version of Wi-Fi.
But now, even WiMax proponents are saying fixed-wireless flavors of the technology are best suited for Third World countries rather than the United States. As for the much-hyped mobile version of WiMax, there is still no standard, and, by the time products appear, it will face stiff competition from emerging third-generation cellular technologies.
The unclear road map and lukewarm commitment from equipment makers raise questions about the viability of WiMax in the enterprise and have industry insiders urging users to be wary of the technology, at least for now.
“Theres a lot of people saying that theyre going to do WiMax, but as we found out with other radio technologies, you have to run the whole race,” said Whitton. “Seven years ago, [on the cellular side] there was TDMA [Time Division Multiple Access], CDMA [Code Division Multiple Access] and GSM [Global System for Mobile Communications], and they all had a path to 3G. Now, its only [variants of CDMA and GSM]. At this stage, being committed means developing prototypes and trying stuff out.”
WiMaxs chief champion is Intel, a company with a good track record of wireless technology publicity. The vendor orchestrated a $300 million marketing campaign for Wi-Fi and the Centrino chip set, which now appears in the majority of Wi-Fi-enabled notebook computers, pushing Wi-Fi into the mainstream lexicon as a must-have technology.
“Were now teaching tens of millions of people a usage model for Wi-Fi,” said Ron Peck, director of business development for WiMax at Intel, in Santa Clara, Calif. “Its the same usage model for WiMax.”
But WiMax isnt Wi-Fi.
Next Page: Standards and Intels marketing muscle could help WiMax succeed.
Standards, Intels Marketing Could
Help WiMax Succeed”>
The road map for WiMax calls initially for wireless, fixed last-mile connectivity and then eventually for mobile broadband connectivity that allows roaming among base stations. The technology promises a range of several miles between client and base station and an average speed of as much as 40M bps per channel. In both fixed and mobile iterations, WiMax is designed to run in licensed spectrum bands, meaning it is contingent on support from telecommunications carriers.
The WiMax Forum—an industry marketing consortium and certification body headed by an Intel marketing director—has largely been successful collecting members. A year ago, the Forum had 47 members; as of April 1, there were 291.
These include telecom equipment makers, wire-line and wireless carriers, and chip-set vendors such as Intel, Fujitsu Microelectronics America Inc. and Texas Instruments Inc. However, membership does not equal commitment to WiMax among mostly pragmatic equipment makers.
“Intel is a big technology partner of Siemens,” said Andy Mattes, president and CEO of Siemens Communications Inc., in Boca Raton, Fla., which makes equipment for wired and wireless carriers and plans to build WiMax equipment. “We are heavily involved, but its too early to make a statement as to where this thing is going.”
Other major Forum members have voiced no product plans. “WiMax is something were tracking closely, but we dont have any specific product plans,” said Dave Leonard, a vice president and general manager of the wireless networking business unit at Cisco Systems Inc., in San Jose, Calif. “Intel kind of expects us to step into the fray.”
The standard for fixed WiMax is IEEE 802.16-2004, also known as 802.16d. The Institute of Electrical and Electronics Engineers ratified the standard last June, and the WiMax Forum plans to begin certification tests for fixed-wireless equipment this July, with hopes for several certified products by the end of the year.
Several base stations have already been submitted for testing, according to Forum officials, who said the success of mobile WiMax depends on the success of fixed WiMax.
“Wheres the credibility factor if you cant deliver good fixed solutions?” asked Ron Resnick, president of the WiMax Forum, also in San Jose. “If you cant do that, why should anyone believe you can deliver good mobile solutions?”
Fixed WiMax, however, is burdened by history. Carriers and equipment providers have been unsuccessful with fixed-wireless efforts. For example, in 2001, AT&T Wireless Services Inc. shuttered its fixed-wireless unit, once dubbed Project Angel. Ciscos acquisition of fixed-wireless player Clarity Wireless Corp. in 1998 also foundered.
“It was a nothing business,” Ciscos Leonard said. “And it was a difficult business to grow. It was a chicken-and-egg thing. Maybe with WiMax, well be able to re-enter that space.”
The existence of a standard combined with Intels marketing push could help WiMax succeed where previous efforts failed, say some industry observers.
“The need for a solid fixed WAN wireless solution definitely exists,” said Daniel Ellis, chief technology officer of PenTeleData, an ISP in Palmerton, Pa. “Many companies, including ourselves, use fixed-wireless links to connect buildings, campuses and such. The issue that we experience is that there are no standards for the licensed professional gear, there is limited interoperability and its expensive.”
“My hope for WiMax relates to rural areas,” said Kevin Wilson, product line manager for desktop hardware at Duke Energy Corp., in Charlotte, N.C., and an eWEEK Corporate Partner. “I have personally talked to a half-dozen people in my company that live out of reach of DSL [and] cable and are searching for connections to work from home.”
Such users could be waiting a while for Forum-certified fixed-WiMax products in the United States, however. The certified WiMax products due to roll out in the fourth quarter will run in the 3.5GHz band, which is not generally used in the United States. The WiMax-appropriate spectrum available in the United States is in the licensed 2.5GHz and license-exempt 5.8GHz bands. Certification testing for equipment using those bands wont happen before next year, the Forums Resnick said.
“Different countries are looking at different frequencies, and thats not going to help deployment,” Siemens Mattes said. “You can adapt to each frequency, but you have to test software a little, and it slows things down.”
Still, some companies are going ahead with public tests. AT&T Corp. plans to hold trials of prestandard WiMax service to enterprises in the United States, including one test late this spring in Middletown, N.J.
For the most part, however, fixed-WiMax proponents are not targeting corporate building links but large rural areas in underdeveloped countries that have little existing wired infrastructure—a substitute for cable or DSL lines.
“I dont expect there to be a whole lot of highly visible 802.16d activity in the States,” said Intels Peck. “Thats my outlook.”
Systems integrators echo the sentiment.
“The biggest WiMax requests are not from the mobile providers but from the wire-line providers asking how do they provide service to rural areas,” said Mattes.
“I more often than not lean to the wired side when mobility is not needed,” said Ellis. “I work for a regional cable/ DSL provider that researches providing fixed and mobile wireless Internet every nine months or so, and each time the business case fails to show a profitable service. Two weeks ago, I attended a town council meeting where Intel presented its boilerplate presentation for WiMax to provide citywide fixed and mobile broadband access for Lewisburg, a small, very digital, already-DSL/cable-wired town in Pennsylvania.”
“The presentation made it fairly clear that WiMaxs strong points for service providers were in areas that lacked existing broadband access—developing countries, rural areas that do not have access and cities that have oversubscribed areas where broadband is not available,” Ellis said.
“Their mobile solution was to install 802.11 [Wi-Fi] hot spots in the areas that needed mobile access,” he said. “As you would expect, the majority of the presentation was all the wonderful things you can do with broadband, but these things are available to the folks already.”
Next Page: Next-generation cellular technologies compete against WiMax.
Next
-Gen Cellular Technologies Compete”>
The mobile version of WiMax, 802.16e, is where the U.S. market lies, according to Intel and Forum officials. But the IEEE has yet to ratify the standard; ratification is expected by years end. Initial products may hit the market in 2007, although many analysts say 2008 is more likely.
“802.16e is going to take a little while,” said Intels Peck. “But were bullish on 802.16e because, while the standard isnt done, its done enough that the silicon starts have occurred. Well see interoperability testing in the second half of 06, and well go from there.”
“802.16e allows for a much bigger play,” said the Forums Resnick. “It means you can have WiMax systems in a portable or mobile-usage model. We believe that long term there could be a concept of personal broadband, people communicating with you via broadband over cell phones. Thats a vision. Its a logical vision.”
Prospective WiMax equipment providers are more cautious about the vision, primarily because of cost.
“It may happen at some point in the future, but I dont think its going to happen in the short term,” said Nortels Whitton. “It took Wi-Fi four or five years before it was inexpensive enough to get into laptops.”
WiMax chip sets are not likely to fall below $20 before 2010, according to a recent study from West Technology Research Solutions LLC, in Mountain View, Calif.
Furthermore, as the industry pushes WiMax along, carriers are also testing next-generation cellular technologies that may well compete against it. Sprint Corp.—which will own most of the 2.5GHz spectrum in the country once it merges with Nextel Communications Inc.—is planning high-speed upgrades to the CDMA-EvDO (Evolution Data Optimized) network. Cingular Wireless, in conjunction with Siemens, is testing a high-speed technology called HSDPA (high-speed downlink packet access), which is related to UMTS (Universal Mobile Telecommunications System).
“While 802.16e promises to be a big market, it is faced with stiff competition from upgrades to existing installed options such as EvDO and eventually UMTS,” said Kirsten West, an analyst at West Technology Research Solutions.
“From the perspective of the Sprints and Verizons [Verizon Communications Inc.] of the world,” West said, “there is little driving them to integrate 802.16e and incur its associated infrastructure costs when the alternative is an upgrade to EvDO that can be done today—and, in fact, is already in trials in several areas across the U.S., with an infrastructure cost that is little more than a software upgrade to the tower in most cases and at worst the addition of a computer board.”
In May, Intel announced plans with Sprint to collaborate on equipment trials and business cases for 802.16e. But Sprint officials warned against counting on WiMax definitively.
“Well probably start to see initial test equipment in the first part of next year, and from there it could be into 07 or 08 before we started to see any mass deployment, but I want to stay away from sticking to any of those time frames,” said Len Barlik, vice president of technology and development at Sprint, in Kansas City, Mo.
“We have not made a decision for a full deployment,” Barlik said. “Were in the investigation phase.”
Thus, some customers believe there are safer bets than WiMax.
“At present, we believe that past failures suggest that a twofold strategy is most prudent—implement 802.11g for our 2 million square feet of in-hospital coverage and use GSM/GPRS [Global System for Mobile Communications/ General Packet Radio Service] for data transfers outside the hospital, hoping that a 3G technology will enable faster bandwidth over time,” said John Halamka, CIO of Harvard Medical School and CareGroup Healthcare System, a Boston-area hospital group.
Check out eWEEK.coms for the latest news, reviews and analysis on mobile and wireless computing.