dailywireless.org » Mobile WiMax

Mobile WiMax - Now?

ABI Research is getting lots of traction on their new research report, Opportunities for High Speed Wireless Data in Portable (802.16e) Markets. The ABI report was widely touted by News Factor, WiFi Planet, Internet News, Mobile Pipeline, Wireless IQ and others.

Northern Sky Research says WiMax hurdles can be easily overcome and predicts that by 2010, there will be 12.4 million WiMAX subscribers worldwide.

On the other hand, Strategy Analytics predicts fixed WiMAX will succeed but questions the viability of mobile WiMax, while E-Week says the fixed market for WiMax is limited and mobilized WiMax is a long way off.

We’re keen to get started on 802.16e, said Mick Reeve, group technology officer for British Telcom. The jury’s still out, of course, but mobile/nomadic could represent a much bigger market.

According to ABI:

Conventional wisdom says that until the advent of 802.16e mobile WiMAX systems still some time in the future the wireless broadband standard will be more or less confined to the great outdoors.
ABI Research analysts say that there are optional specifications built into the 802.16 standard which can boost the sensitivity of receiving equipment to the extent of making WiMAX PC cards and built-in receivers a practical proposition for laptops, PDAs and other portable devices. Generally these optional specifications have not been implemented by the largest vendors of WiMAX equipment.
However at least two smaller companies TeleCIS and Sequans have been designing their chipsets to implement these under-utilized options in the standard.
According to senior analyst Philip Solis, what this means is that “There may be WiMAX PC cards on the market earlier than many observers have expected.
These will result from superior chipsets permitting the use of WiMAX in laptops and similar devices in homes and offices within the reach of fixed WiMAX transmissions. You will not have full mobility as you will with 802.16e, but you will have some portability.”

Well, yeah. TeleCIS and Sequans have been promoting mobilized versions of WiBro/WiMax, using gate arrays. And they’re working on client chips for mobilized WiMax/WiBro. Who isn’t? TeleCIS says the bill of materials for a dual-antenna (MIMO) CPE is expected to be under $100 for a $200-$300 user price with better performance. TeleCIS claims their MIMO-like design provides 12 to 17 dB of additional gain. But it’s a year or two off.

The Intel/LG deal blends elements of WiBro/WiMax. WiBro offers the speed of WiFi with the range of cellular. It provides throughput of 30-50Mbps with a 1-5km (.5-3 mile) range (at 2.3 GHz) - about like cellular, but faster and probably cheaper, with IP all the way. In the United States, it all hinges on Sprint and Clearwire - they own most of the usable 2.5 GHz frequencies. The 5.8 GHz unlicensed band is okay for backhaul - not for mobility. Sprint is unlikely to undercut their EV-DO service. That leaves Craig McCaw’s Clearwire.

Broadband Policy in the United States

Broadband wireless policy for the United States may be largely determined by Craig McCaw (and maybe SK Telecom). Kevin Martin will do as he’s told. If Verizon/SBC/Comcast manage to sink municipal broadband or Clearwire, Powell left an out - the unlicensed 700 Mhz TV band (via 802.22). It’s not perfect, but it’s a workable plan.

Scalable-COFDM is the new thing at WiMax HQ. WiMax is changing horses mid-stream with S-COFDM. It promises improved coverage with mobility, but at the cost of backward compatibility with 802.16-2004. But since there really aren’t any WiMax clients out there, the WiMax Forum is hoping nobody will notice. Meanwhile, Intel, Fujitsu and others have modified their chips. They expect S-COFDM will be standard issue in the future for CPEs and basestations, largely supplanting 802.16-2004 even before it gets off the test bench at Cetecom.

802.16-2004: Lost In Translation?

The 802.16-2004 got a range boost using subchannelization. That allows clients to use fewer carriers (upstream) resulting in higher EIRP. That helps indoor clients get out.

A scalable carrier is the next frontier for Mobile WiMax. But moving to mobilized WiMax (and S-COFDM) may not be the smooth transition everyone was hoping for. Plain vanilla 802.16-2004 isn’t compatible with the new Mobile WiMax/WiBro standard. That’s forcing chip companies to create “enhanced” 802.16-2004 WiMax chips, with a foot in each camp.

Two mobile standards (IEEE 802.16e & WiBro) agreed to merge to avoid political, technical and economic headaches. Intel wanted the Scalable COFDM technology while South Korea wanted early entree into the global WiMax market. Intel and LG are now working together to bring Korea’s 2.3 GHz WiBro system to WiMax. Elements of WiBro, specifically scalable COFDM, are being incorporated into Intel chips.

Samsung and LG Electronics lead the world in WiBro - and may have first mover advantage in Mobile WiMax. Intel couldn’t afford to loose China. South Korea couldn’t afford to loose a global market.

Lost in translation may be IEEE 802.20 and the original IEEE 802.16-2004 standard. The mobilized IEEE 802.20 “standard” now appears to offer few advantages over the WiBro/WiMax “e” juggernaut. Cellular support is tepid as EV-DO and HSPDA loom. Flarion is promoting their FLASH-OFDM system as the “defacto” standard.

Meanwhile, “enhanced” WiMax chips, with the ability to interoperate with Scalable COFDM (for better range), may mean the plain vanilla 802.16-2004 standard could be quickly obsolete.

At least that’s the way we see it at DailyWireless. But we may have it all wrong.

Intel and LG are working together on WiBro/WiMax while TeleCIS Wireless, a California chip designer, has strategic alliances with Samsung and Korea Telecom. Meanwhile Unstrung reports that LG Electronics will use Seattle-based Adaptix channel cards in LG’s WiBro products for basestations.

But ABI’s press release seems a bit confusing. The TeleCIS gate-array solution is not the same as an “enhanced” client. Secondly, “enhanced” WiMax clients may be fairly common. Mainstream chip providers like Intel and Fujitsu are apparently getting “enhanced”, too.

Both KT, South Korea s largest fixed-line carrier, and SK Telecom, Korea s largest mobile operator, hope to have operational WiBro service early next year. SK Telecom expects to have a dual-band CDMA 2000 1x and WiBro handset available at the time of commercial launch next year.

Meanwhile, Sequans, the French WiMax startup, is eyeing Korea’s WiBro developments closely, according to WiFiPlanet:

Sequans is initially focusing on fixed WiMax (802.16-2004), with the aim of working towards mobile 802.16e as quickly as possible as the standard matures. “Our mobile development is initially targeted towards the Korean space which is WiBro because we believe that ultimately WiBro is going to be the first deployment of mobile WiMax applications,” says Bernard Aboussouan, Sequans’ vice president of Marketing.

Sequans chips work in both base stations and subscriber units. It’s currently shipping “pre-WiMax” FPGA boards to customers. “By the middle of this year, we’ll have our system-on-a-chip, an ASIC, and by the end of this year we plan to have an 802.16e and WiBro chip for the mobility space,” says Aboussouan.

Airspan is using picoChip’s PC102 processor in its WiMax basestations. A software upgrade converts them from the fixed to the mobile WiMAX standard, enabling the basestations to connect to laptops, PDAs and other mobile devices that will be supported by the emerging 802.16e mobile WiMAX standard. The basestations will be commercially available in the third quarter of 2005. Airspan clients use Intel’s WiMax chips.

As Intel explains, neither of the two WiMax standard modes; the 256 carrier (OFDM version) or the OFDMA (Multiple Access - without scalability), could deliver vehicular mobility at high speed. Furthermore, WiBro’s scaleable carrier system could accomodate bandwidths from 1.25 Mhz to 20 Mhz easily.

Meanwhile, today the Infocomm Development Authority of Singapore (IDA) said six companies successfully bid for Wireless Broadband Access (WBA) spectrum in Singapore. Bidders can start deploying their WBA networks from 1 July 2005, and have to offer services within 18 months. Bidders included Pacific Internet Limited, the largest telco-independent ISP in the Asia Pacific region by geographic reach, secured 30MHz of frequency spectrum in the 2.5GHz band. That’s supported by WiMAX. The winning bidders will deploy wireless broadband nationwide in Singapore.

A contrarian view is held by Strategy Analytics. They think power consumption will limit Mobile WiMax compared to the established base of mobile 3G users.

E-Week says the fixed market for WiMax is limited and mobilized WiMax is a long way off:

“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”. “[802.16e] may happen at some point in the future, but I don’t think it’s going to happen in the short term,” said Nortel’s Whitton. “It took Wi-Fi four or five years before it was inexpensive enough to get into laptops.”

Meanwhile, RemotePipes has formed a WiMAX Global Roaming Alliance (WGRA) that aims to develop roaming interoperability standards for future WiMAX broadband wireless networks.

Concerns of the Alliance will include streamlining authentication and accounting functions between network operators, as well as ensuring uniform user processes between networks.