New Directions for Broadband Wireless

Three renegades say their technologies can overtake 3G.

Remember the promises of third-generation wireless technologies? Mobile workers with secure, 24-by-7 broadband access to the companys servers via wireless modems? Live videoconferences on cell phones? Opportunities to develop streaming video and audio products for handhelds? Seamless, transparent access to an invisible network as pervasive as the air we breathe?

In IT departments, the term 3G has become synonymous with vaporware. However, things are about to get a lot more interesting in the wireless data market. A handful of well-financed upstarts see 3Gs setbacks as an opportunity to push the virtues of their competing high-speed mobile data technologies. Its far too early to predict the outcome of this competition, but the ultimate winner is likely to be the mobile worker.

Three companies—Broadstorm Telecommunications Inc., Flarion Technologies Inc. and Tantivy Communications Inc.—are in various stages of developing all-IP wireless data technology that they said will deliver the promises of 3G without many of its costs and hassles. The challengers offer a different vision of the wireless future, one they said is more likely to meet the needs of data users in general and is especially suited to enterprises already using 802.11b wireless networks. In their no-3G scenario, mobile workers will access data at speeds up to 1M bps each over the air using standard laptops and handheld devices. Even more important, they will use the same Internet applications they now use on their tethered hardware. But none of this is a sure thing. These companies will have to fight powerful vendors, global standards and legacy networks while struggling to survive todays tough economy.

Still, the current environment could prove to be fertile ground for strong competition to satisfy a growing appetite for wireless data delivery. Thats because wireless carriers have found implementing 3G to be astoundingly expensive, and the technology has so far failed to deliver promised data speeds.

Dave Nowicki, vice president of product management and marketing for ArrayComm Inc., of San Jose, Calif., which has developed a wireless data network technology, calls 3G "a terrific vision." Nevertheless, Nowicki said, "many results of 2.5G and 3G technologies today havent lived up to the expectations of that vision."

European operators that pioneered 3G networks are finding them so expensive that some have been forced into sharing networks. In France, the government has had to reduce payments for operators and extend the payback period. And in Japan, NTT DoCoMo Inc., the first operator to commercially launch 3G after a long-delayed rollout, is having trouble marketing an expensive service that has largely failed to satisfy users.

There have been many reasons for these setbacks, not the least of which is that 3G is built on nonstandard IP implementations, requiring that Web services and applications be customized for access via tiny-screened, wireless handheld devices. The challengers, in contrast, have all made their products fully compatible with standard IP.

Against that background, the challengers are gambling that service providers may be ready to rethink their commitments to 3G.

"Some operators are thinking along the lines of skipping 3G because 3G spectrum costs are outrageous, and the upgrade cost is outrageous," said Michael Schy, director of marketing and business development for Broadstorm, of Bellevue, Wash.

Schy and his counterparts at Flarion, in Bedminster, N.J., and Tantivy, of Melbourne, Fla., said their solutions deliver data far more efficiently and less expensively than 3G technologies. Incorporating IP end to end, they deliver from 365K bps to 8M bps per user. That compares favorably with the highest network speed achievable with the 3G standard, CDMA2000 1xEVDO, which offers a 2.4M-bps peak data rate that is shared among users.

The challengers migration paths from circuit-switched to packet-switched offerings can also be attractive. For example, Flarions technology requires a dedicated chunk of the carriers spectrum, but by offering voice over IP on the data portion of its network, the carrier can migrate from voice to data as the market evolves, eventually phasing out circuit-switched networks. That concept of using separate spectrums is similar to the idea behind CDMA2000 1xEVDO, the 3G step in the evolution of Code Division Multiple Access cellular networks beyond the CDMA2000 1xRTT standard, which requires a separate spectrum channel for a data-only service.

The non-3G schemes share a relatively simple network architecture: an antenna, which in many cases can be shared with the voice network, and a radio that interfaces via an IP router to the Internet. In contrast, the CDMA2000 1xRTT and WCDMA (Wideband CDMA) standards require extra components to enable the voice network to communicate with the IP-based Internet, adding expense and complexity to a build-out.

Most of the 3G challengers were born from the desire to build efficient wireless data networks unencumbered by the complications and expense of accommodating a legacy infrastructure. Flarion was spun off from Lucent Technologies Inc. to solve this issue. "There was no consideration of being backward-compatible with anything legacy," said Rajiv Laroia, a Lucent veteran and founder and chief technology officer of Flarion. Laroia was charged with designing the ideal wireless data network.

Because their technologies are built on standard IP, Broadstorm, Flarion and Tantivy require only that operators hook up the wireless base station to an IP data line that connects to the Internet. "All we want is an off-the-shelf IP network with off-the-shelf routers," Flarions Laroia said.

That simplicity frees operators to choose equipment such as PacketAirs Mobility Router and back-end gear from Cisco Systems Inc. and other vendors that best satisfy their needs rather than being locked into proprietary equipment. PacketAir Networks Inc., of San Diego, a developer of packet-based wireless networks, specializes in building routers that fit between an IP radio base station, such as those built by Broadstorm, Flarion and Tantivy, and the Internet and the company is touting its products for the new network architectures.

Currently, although 3G is a standard, equipment from different suppliers doesnt interoperate. "If you choose Ericsson AB for your over-the-air equipment, you cant buy Nokia Corp.s back-end equipment because they dont interoperate. Thats been the game," said Matt Brookshier, Packet-Airs vice president of marketing. An all-IP architecture promises to open up competition and push down prices dramatically, Brookshier said.

But even if all 3G routers and back-end gear were compatible, a major drawback to 3G is that, unlike other data pipes, it makes up its own rules for IP. Since the challengers all base their technologies on IP, any Internet application can be accessed over their networks. Especially for enterprise customers, this promises a substantial savings over 3G because it allows mobile workers to access Web applications and services wirelessly—a major time and cost savings for IT departments. And because these networks promise significant capacity and efficiency, enterprise users can also expect to pay carriers flat, all-you-can-eat fees, similar to land-line Internet connections.

Broadstorm, Flarion and Tantivy use different air interfaces than those employed by wireless carriers but said their products are more efficient. Tantivy calls its technology I-CDMA, which uses CDMA but doesnt rely on intellectual property owned by Qualcomm Inc., which holds numerous patents on the CDMA standard widely used in the United States.

Because its technology is built on CDMA, Tantivy can incorporate all the same radio-frequency components— among the most costly pieces of a wireless network—used by the existing CDMA systems.

Tantivys pipe can carry 1.9M bps, which typically translates to about 365K bps per user on the downlink and 200K bps on the uplink. The high-speed uplink is a differentiator for Tantivy and would enable bandwidth-hungry applications such as live mobile videoconferencing. Tantivy said it will sign service-level agreements guaranteeing that 90 percent of a networks users at any given moment will get at least 256K bps throughput.

In addition, Tantivys technology can penetrate more walls than 3Gs CDMA2000 1xEVDO, said Tom Gorsuch, vice president of product development and co-founder of Tantivy. It also allows Tantivy to deliver high data rates into a larger portion of the cell coverage area, Gorsuch said. Tantivy also claims its network can support five times as many users as CDMA2000 1xEVDO—up to 1,000 customers per base station.

Broadstorm and Flarion employ the OFDM (Orthogonal Frequency Division Multiplexing) air interface. "OFDM is probably the best way to do point to point, and people realize that, but they couldnt make it work with multiple users, especially in a mobile environment," Broadstorms Schy said. "Weve been able to do that." He said Broadstorm can deliver a maximum of 8M bps.

Flarion has developed its own flavor, called flash-OFDM, which Laroia said is more efficient than existing standards because it is tone-hopped spread spectrum. Standard CDMA is direct spread spectrum and employs a scheme to apportion audio spectrum that is far less efficient than OFDMs use of tones, Laroia said. As a result, while CDMA users communicating with the same base station interfere with one another, OFDM apportions tones in a way that prevents such interference. Laroia said this enables Flarions technology to handle three times more data than CDMA can handle.

"I truly believe that Flarions flash-OFDM is superior to the other technologies," said Phil Marshall, an analyst with the Yankee Group Inc., of Boston. But he also views performance claims skeptically. "We dont have live networks to validate the performance," Marshall said.

Customers would typically have a choice of service levels with any of the challenger technologies because Broadstorm, Flarion and Tantivy are all building QOS (quality-of-service) capabilities into their products. Tantivys offering will let users log into a billing system and self-provision QOS. For example, an IT manager could order a higher level of service for a limited time period to conduct a videoconference, paying for the higher bandwidth only when it is needed.

Sound too good to be true? Critics say the emerging challenges to 3G may be just that.

Qualcomm and some vendors of 3G technologies insist their solution is as good as, if not better than, those touted by the challengers. For example, they pointed out, CDMA2000 1xEVDO has been accepted by the United Nations International Telecommunications Union as a standard. As such, it has a greater chance of achieving economies of scale in equipment prices. It is also backward compatible—and tightly integrated—with existing CDMA networks.

To cite just one example, Airvana Inc., a developer of CDMA2000 1xEVDO network products, recently cut a joint development and supply deal with Nortel Networks Ltd. under which Nortel will incorporate Airvanas gear into its future products. Existing Nortel customers will be able to upgrade their networks for CDMA2000 1xEVDO with Airvana plug-in channel cards.

Airvana, of Chelmsford, Mass., said users will get 700K bps to 1.2M bps on the downlink and 153K bps on the uplink. Qualcomm, of San Diego, said another advantage to building on existing voice networks is that operators can continue to use their installed back-end systems for billing and provisioning.

While the challengers argued that their proposed architectures are less expensive to deploy than 3G gear, each employs proprietary or little-used air interfaces, which increases production costs. While the OFDM interface used by Flarion and Broadstorm is getting more attention in fixed wireless broadband systems, its not as widely used as the current mobile wireless standards. As a result, the cost to produce OFDM-based products or other proprietary technology will be high, at least initially.

In addition, Flarion and Broadstorm are asking a wireless market that has waged fierce, often irrational battles over air interface technologies for significant changes in the mind-sets of carriers already entrenched in either the CDMA or GSM (Global System for Mobile Communications) camps.

That means that enterprises interested in one of the new technologies might also have to deal with an upstart carrier because non-3G data pipes would be most attractive to a new operator rather than an existing player that already has huge investments in gear.

Those new operators, and the upstarts that hope to serve them, may get their chance, though, if those committed to 3G cant make good on its promise soon. Theres only so much time that any technology can remain in vapor before the IT community says, enough is enough.

Nancy Gohring is based in Seattle. She can be reached at [email protected]