The mobile WiMAX standard, 802.16e, has finally been frozen, the precursor to imminent ratification by the IEEE. This will then pave the way for the start of the testing and certification process and the arrival of standards-assured gear, the latter expected towards the end of 2006. In the year-long hiatus before that happens, pre-standard technologies offering mobility and a migration route to 802.16e will be fighting to gain presence and so put themselves in a strong position in the competitive market that is sure to develop around mobile WiMAX from 2007. One of the best placed, of course, is the Korean technology Wi-Bro, on which 802.16e is largely based, and which has the advantage of big vendor backing, primarily from Samsung, and guaranteed customers in the shape of the two major Korean operators, KT and SKT, which are rolling out Wi-Bro services next year.
The convergence of Wi-Bro and 801.16e is the best opportunity to date for Samsung to amass a global market for its infrastructure technology (see Wireless Watch 2 September 2005), and to usurp Intel’s self-created role as driver of broadband wireless standards and devices. This bid received its greatest credibility boost to date this week, with news that Sprint Nextel had signed the Korean giant to trial Wi-Bro technologies in its 2.5GHz spectrum, which covers 80% of the US market and in which it is mandated to build a major broadband wireless service, reaching 30m by 2009. The agreement will involve testing prototype terminals and supporting core network equipment designed to support interactive multimedia services. The collaboration will involve trials at company facilities and field trials with customers, as well as interoperability testing with Sprint Nextel’s ecosystem partners. The partners will test access devices including a wireless air card and integrated handset, although the handset may not be ready at the start of the trial period, which begins in October. Target data rates are 3Mbps download and 1Mbps upload on average.
A week earlier, Sprint Nextel had agreed to trial equipment from Motorola, one of the majors with the most aggressive commitment to lead the mobile WiMAX market, but now it seems the US suppliers will face a challenge from less familiar Korean challengers.
If Samsung can demonstrate that its pre-802.16e technology can perform effectively in 2.5GHz bands under US regulations, it will quieten criticisms that Wi-Bro is suited mainly to the 2.3GHz bands and more generous allocations of spectrum of its native country. It will not only be able to point to having the first really large carrier deployments of pre-802.16e services in Korea, and all the valuable experience that should be gained from the early activities of KT and SKT, but will be able to claim an internationally applicable solution. Its moves will also probably accelerate the appearance on the market of certified 802.16e gear for the 2.5GHz band. In the fixed WiMAX testing schedule, this profile takes third place to 3.5GHz and 5.8GHz, but the importance of the Sprint Nextel deal ‘ as currently the largest concrete WiMAX plan in the world, thanks to the roll-out stipulations set by the FCC as a condition of the company’s recent merger ‘ should push 2.5GHz up the agenda in the mobile certification process, especially as the 2GHz bands will be the best suited of the current WiMAX profiles for mobility.
All this backs up the conculsion of a new report from ABI Research, which predicts that Wi-Bro will not only flourish in the Korean 2.3GHz bands for which it was originally designed, but because it is now also the precursor to 802.16e, will gain traction outside its home country. The longer the wait for full ‘e’ products, the greater foothold Wi-Bro could gain, stealing a march on other pre-e products such as Navini Ripwave or Motorola Wi4.
Before Wi-Bro and WiMAX camps agreed to harmonize their technologies, Wi-Bro was widely regarded as a local specification fulfilling Korean local conditions only, but Samsung has grander plans for it than that, and in July, the Japanese government accepted Wi-Bro, along with 802.16e, as a next generation broadband wireless standard candidate.
This will make alliances between Samsung and western majors increasingly likely. The Korean company may recently have thrown down the gauntlet to Intel, claiming it will prevent ‘Intel Inside’ being universal in next generation devices, but it cannot establish a global customer base, especially in WiMAX, without cooperating with the chip giant ‘ just as Intel recognized last year, despite earlier efforts, that it could not exclude the Korean majors from the mobile WiMAX platform, for fear of dividing the world around two broadband wireless standards.
Similarly, Motorola and Samsung may meet head-on in the US market in the network as well as the handset market at last, but they are also likely, away from the direct selling process, to have considerable cooperation in the Sprint Nextel trial ‘ which also involves Intel, through its major handset-oriented R&D venture with the operator, which is also likely to take in the chipmaker’s R&D alliance with Nokia. So the complex network of alliances will be created that will be necessary to transform WiMAX from an interesting new technology to a carrier class, real world platform with a full ecosystem and vital features such as attractive handsets.
One interesting alliance that demonstrates the strengthening R&D lines of communications between the west and Korea is between PicoChip, the UK base station chip designer that provides Intel’s forthcoming WiMAX base station architecture, and ETRI, the Korean government-backed technology research institute, under whose auspices Wi-Bro was devised, and which has a profound influence on the country’s fabled wireless and broadband developments. The new partners will work on software defined radios for WiMAX as well as for 3G (W-CDMA and HSDPA). PicoChip’s early work was focused on multi-network, software defined base stations capable of supporting two or more wireless platforms simultaneously, and as the market starts to demand such capabilities, we can expect the company, and its partner Intel, to be prominent in the development of dual-mode base stations for companies wanting to support 3G and WiMAX within one structure.
Dr Yim Chu-Hwan, president of ETRI, said: “For our software defined radio project we evaluated all the programmable architectures and selected PicoChip on the basis of price/performance, flexibility and suitability for next generation wireless technologies’, as well as processing power considerations.
PicoChip has also announced a family of reference designs based on its picoArray processor, which now covers both subscriber station and base station designs for fixed and mobile WiMAX as well as Wi-Bro and local variants proposed for China. PicoChip claims to be the first company to offer all these options within one platform, with software upgrade promised between different versions of the standard, and also to include emerging features for advanced WiMAX systems, such as smart antenna arrays. As well as Intel, PicoChip’s major partner is Airspan, whose equipment is OEMd by Ericsson, Nortel, Fujitsu Networks and Marconi.
In signing up Nortel, Airspan has secured one of the last equipment majors not to have formed an OEM deal or announced its own WiMAX roadmap. The agreement shows the importance of Intel in its role as the spider holding together the web of alliances surrounding its approach to WiMAX ‘ a web that should, like Samsung’s parallel ecosystem, speed up evolution of the technology. So Nortel has a development deal with Intel focused on customer devices ‘ which may be able to tap into the important handset alliance with Nokia ‘ and uses the Airspan equipment, which in turn will be based on the Intel platforms, Rosedale at the client end and Intel/PicoChip for the base station. Nortel also brings into the mix its own joint development agreement with Korea’s LG, which includes broadband wireless and Wi-Bro in its remit as well as W-CDMA, and is another strong example of strengthening Korean-North American R&D links (LG is working with software defined radio specialist Adaptix on a Wi-Bro/WiMAX platform that will compete with PicoChip’s).
Nortel will work in the development of WiMAX/Wi-Bro solutions through its LG partnership. Its fixed WiMAX offerings will be available in the first quarter of next year and its Wi-Bro, pre-802.16e solutions in mid-2006, following a field trial this year in Korea and one next year in North America.
It is likely that Nortel will eventually develop its own mobile WiMAX kit to replace the OEM deal, in common with rivals such as Alcatel – which plans to shift from OEMd equipment to homegrown 802.16e as the market develops. But Nortel, although heavily focused on Wi-Bro developments, is specific that it will use Airspan kit for both fixed and mobile WiMAX, targeting, in particular, T1 replacement and cellular backhaul with the former. This is a similar line to that taken by arch-rival Lucent, which OEMs Alvarion fixed WiMAX gear and has now said it will also extend this deal to the mobile market, at least in the early stages. Such decisions will be crucial for the WiMAX specialist vendors, which were facing crisis if all the majors decided to use licensing deals only for the fixed market and create their own mobile offerings.
“We are combining Nortel’s wireless and mesh technology leadership and ability to execute large network build-outs with Airspan’s market proven WiMAX portfolio,’ said the company. “Beyond that, Nortel will build on its key intellectual property and standards contributions in OFDM, MIMO, and wireless mesh networking to deliver mobile WiMAX products in time for trials in mid 2006. The key to an operator’s success in WiMAX will be the real world implementation of networks, an area where Nortel has vast experience and the resources to back it up.” Nortel’s optical backhaul technology, data networking solutions, and suite of wireless services will be integrated with Airspan’s base transceiver stations and subscriber units.
Indications are that Nortel will stick with Airspan at least for the first couple of generations of kit, probably until the market is large enough to justify a self-developed product, and until it evolves technologically to the stage where Nortel can have a significant influence and competitive edge based on its own R&D. It certainly has very advanced R&D projects in broadband wireless, spanning cellular technologies (see separate item on CDMA), Wi-Fi and future OFDM-based technologies (see Wireless Watch May 26 2005). In March 2004, for instance, it completed a MIMO/OFDM demonstration of large data transfers with peak data rates at 37Mbps over a standard 5MHz PCS band.
As such, it will make a major contribution in its own right to the Intel ecosystem and the broader WiMAX community. Its own financial position remains unsteady but it is creating one of the strongest and most complete multi-network strategies, and amassing serious intellectual property and technological advantage. Whether it has the strength and stability to exploit this to the full is a bigger question, and we still believe these innovations could have a greater impact on the market if Nortel were to be acquired, or form a very close venture with, one of its rumored suitors from the integration and IP worlds, such as Cisco or, more attractively still, its IMS ally IBM.
The growing intimacy of the large North American vendors and chipmakers, and the Korean trailblazers, confirms the position of Samsung and LG as the technological agenda setters for early mobile WiMAX, but despite political tensions, also shows an unusual level of cooperation between giants from both sides of the world, to accelerate bringing their chosen technology to reality before it can be eclipsed by the advances of CDMA and HSDPA. This means that Wi-Bro will certainly be the most important pre-802.16e variant in licensed spectrum, in the US as well as Korea, but this does not guarantee Samsung’s long term success as a vendor in this sector, despite its intellectual property advantages. The multi-network suppliers with their focus on convergence and their close contacts with the major telcos ‘ including Lucent, Nortel and Motorola ‘ are all positioning themselves strongly too, and are happy to spread their favors evenly between the Korean majors and Intel in terms of pushing the R&D agenda.