One of the most promising markets for WiMAX, especially from 2010, is as a data overlay for GSM operators, in areas where there are no 3G licenses as yet, or where 3G will not be cost effective. Phonemaker HTC is pointing the way, unveiling a GSM/802.16e dual-mode device that will initially be offered by Russian operator Scartel, which has an ambitious 2.5GHz build-out program, starting in Moscow and St Petersburg.

Operating under the brand name Yota, Scartel has carried out trials, using Samsung infrastructure, in the two major cities, and says it will go commercial by year end, having built out 1,600 base stations at a cost of $200m by then. Emulating Sprint’s Xohm launch, it says it is essential to offer a range of mobile broadband terminals from day one, to encourage high value customers such as business travellers - and these will make roll-out of fixed services, especially in more rural areas, a more viable proposition.

As well as the HTC handset, Scartel will launch with USB modems from Samsung and Asus,  plus ExpressCards for PCs and, from early next year, it hopes to add support for multiple access WiMAX/Wi-Fi embedded laptops and dongles.
 
The handset is the star of the show, though, supporting wide area roaming on Russia’s extensive 2G network, not just on Wi-Fi hotspots, which are sketchy in coverage outside the metro areas. The phone, called Max 4G, will automatically switch to a VoIP-over-WiMAX connection, at low rates, when two Yota subscribers are connected to one another within the 802.16e coverage zones.

The handset is a good example of how WiMAX operators in emerging but high growth markets need to balance their appeal to two key constituencies - the bulk of the target base will require low ARPUs and basic access services, but there is a sizeable and growing population of business travellers and other high value subscribers. The latter will want not just basic cellphones, but devices with all the features expected by smartphone users in the developed mobile economies. Hence, the Max 4G will feature a 3.8-inch 800 x 480 touchscreen display, integrated GPS and Wi-Fi, an accelerometer, a 5-megapixel camera and will use the Windows Mobile 6.1 operating system.

Yota’s basic subscription allows customers to access online games, maps, messaging and file exchange applications while on the move, and there will be add-on services for accessing films, video and TV programmes online.

The non-branded version of the device, which is basically a variant of the popular HTC Touch midrange smartphone, with WiMAX added, is called HTC T8290 and will be offered to other WiMAX operators. A CDMA model is expected by many to be developed, so that it could be targeted at Sprint/Xohm and Japan’s KDDI/UQ venture. There is also much speculation that HTC will create an Android handset for WiMAX, similar to its G1 phone for T-Mobile’s 3G network. Android, once better established, is likely to prove a more mainstream option for developing economies than Windows Mobile, which remains largely the preserve of the mobile enterprise user.

And there is a large potential market for GSM/WiMAX combinations, with some chipmakers, like Comsys, already heavily focused on this sector. In recent research by Rethink Technology, a significant base of operators was identified, holding only 2G networks, but needing to increase their revenues and customer retention over the next few years by adding data, internet and multimedia services. The research calculated that, in the global base of 2G-only carriers, almost 200 will be using WiMAX in some portion of their footprint by 2013, in a multimode system combined with GSM/EDGE and sometimes 3G. In capex terms, mobile operators will account for about one-third of Mobile WiMAX spending by 2012 and the cellular/WiMAX base is the largest market for 802.16e terminals by 2012, and the second largest market (after DSL alternative and fixed operator convergence) for infrastructure.

From HTC’s point of view, creating handsets for operators in this situation, as well as for carriers in developed economies like Xohm, would be in line with its strategy of stealing a march on larger players by supporting minority technologies at an early stage. It was the first cellphone manufacturer outside the pure ODM community to create Windows handsets, using them to establish a high end range under its own brand, rather than focusing entirely on white label devices, its original business model. It remains willing to submerge its brand in order to gain early presence in a new and promising market, ahead of the competition, as it has done with T-Mobile G1.

Given Xohm’s devotion to open access, a WiMAX Android phone would not only enable HTC to be a first mover again, but also to launch an Android product under its own name in the future. Android products for Xohm are widely expected for mid-2009, given that Google and Sprint Nextel have collaborated closely on the software platform and user experience, and Google is a major supporter of the open access web services model on which Xohm is pinning its hopes. HTC will surely be positioning itself for that opportunity, as well as for the higher volumes promised by the GSM world.

In the developed mobile world, WiMAX will have to live alongside several other technologies, and consumers will increasingly want to be able to roam seamlessly across different networks according to cost and availability - especially if open access models start to take off in earnest. This will place new momentum behind the bid to make multimode devices more complex, and yet cheaper and less power hungry.

Packing more and more radios on to a single chip is one approach, as pursued by Qualcomm (though no support for WiMAX yet). A stage beyond that comes the software programmable processor, which can be tweaked in software, by the handset maker or operator, to support a flexible combination of network connections. Sandbridge Technologies this week announced its contribution to this nascent market, with its SB3500, a 65nm reprogrammable software-based baseband processor, which can be adapted to LTE, HSPA, WiMAX, W-CDMA, Wi-Fi, GPS and the mobile TV standard DVB-H.

The processor is entirely software-based, and points to the vision of a fully cognitive radio in future - a single radio that can move intelligently between different networks, according to parameters such as cost or signal strength.

Because the platform is programmable, cellphone makers can use a single architecture to support phones with different combinations of radios, reducing their costs. Sandbridge’s technology will also reduce the number of chips needed in a phone, which should cut cost, time to market and complexity. The company claims the cost savings can amount to 15% across design and production, and the figure could grow as consumers look for more and more functions to be packed into one low power unit.

Another advantage that Sandbridge claims for its technology is that handset makers can respond more quickly to standards changes - such as the next wave of 802.16e, or even the shift to 802.16m - testing and adding new functions rapidly in software rather than waiting for the chipmaker to deliver upgraded hardware. Changes in standards can also be easily accommodated, so operators could move to new networks even before standards are finalized, with reduced risk.

As for competition, there are other start-ups, such as Sweden’s Coresonic, in this field, and ST-NXP Wireless has some advanced software defined technology, from the NXP side of the group, but most of the majors are nowhere near to market readiness with programmable basebands yet.

As for more conventional multimode devices, ABI Research, contrary to many market assumptions, believes many carriers will adopt a mixture of the two OFDM-based systems, which share many similarities. The company forecasts that ­2009 will see the introduction of dual-mode WiMAX/LTE chipsets for devices.

“Some mobile operators are showing interest in dual-mode chipsets,” said ABI principal analyst Philip Solis. “And they are backing it with cash. Vodafone, for example, has a foot in both WiMAX and LTE camps. They will use LTE in industrialized regions, and WiMAX in developing nations. In Japan, KDDI may deploy LTE on its own, but as an investor (along with Intel and others) in WiMAX operator UQ Communications, KDDI has an interest in both standards.”

Looking ahead to full cognitive radios - which actually sense wireless traffic to make intelligent decisions about using spectrum - Intel hosted a workshop last week in conjunction with European research institute IMEC. The chip giant believes the FCC’s recent decision to open up the white spaces in broadcast spectrum will light a new fire under cognitive radio work and offer WiMAX a new opportunity. Some early work on interference avoidance in the spaces was undertaken by 802.16 study groups.

Among companies represented at the workshop were Alcatel-Lucent, Alvarion, Marvell, Motorola, Nokia and Qualcomm and IMEC aims to build prototype hardware soon, geared to next generation WiMAX and LTE networks in licensed bands, as well as to various technologies for license-exempt white spaces.

Cognitive radios would “have more direct pay-off to the incumbent operators and some form of this probably will begin to appear in next generation broadband standards such as WiMAX 2 [802.16m] and LTE,” said Kevin Kahn, director of Intel’s communications technology lab. He also believes the white spaces will be a good area to test the new technology at relatively low cost and risk.

Today, November 4, is not just a crucial date for US politics, but also for WiMAX, as the FCC prepares to vote on the merger of Clearwire with Sprint’s Xohm unit to form what should be the first operator to offer true wireless broadband on a national scale. Assuming the vote is positive, the deal should be finalized by year end, and will represent a turning point in the US wireless industry. But of course, ‘nationwide service’ rarely means that in the US, so another, less publicized, FCC decision, also announced this week, could be highly significant too, for rural US citizens and for the WiMAX platform. This is the clearance for mobile satellite giant Globalstar to run WiMAX in its spectrum, creating hybrid satellite/terrestrial networks that are cost effective for rural coverage, and could bring mobile broadband to the people that Clearwire misses.

Clearwire CEO Ben Wolff said last week that he was confident that the FCC vote would go in favour of the Sprint venture, with no significant conditions. “Everyone understands it’s a pro-competitive deal,” Wolff said. “It creates a new nationwide broadband network. It’s not a partisan issue. It’s just good policy.” If the FCC votes yes, Clearwire shareholders will vote on November 19, the last major hurdle to cross, and the transaction could then be completed by year end as planned. The main objections have come from AT&T, and some other telcos, which argue the deal, and especially the huge spectrum base it concentrates in the hands of one operator, should be scrutinized more closely. However, the FCC has been keen to foster new services in wireless broadband and is likely to see the enlarged Clearwire as a new entrant bringing new consumer choice - and so will favour the merger that will strengthen its chances of undertaking its roll-out and being commercially viable. The deal involves a $3.2bn capital injection from Intel, Google and three cablecos (Comcast, Time Warner Cable and New House). Both halves of the deal, Xohm and Clearwire, are rolling out Mobile WiMAX networks, but clearly their business model hardly makes sense without the increased capital of the new stakeholders. While legal requirements mean they have to appear to be acting very separately, Wolff said in his interview that they were coordinating as closely as the law allowed, so as to gain maximum efficiencies once they become a single entity. He said: “If we have this deal closed by the end of the year, it would clearly make sense for us to have everything from back office preparation to branding and marketing on the same page.” Wolff will remain CEO of Clearwire after the deal closes, while Xohm chief Barry West will become president. Despite the new investors, the company is expected to seek to raise an additional $2bn-$2.3bn in funding fairly quickly, global markets allowing - the $3.2bn infusion will take it through to 2010. One huge problem for any wireless business model is the difficulty of making a profit from rural networks, built in areas of sparse population - especially when the carrier is working in 2.5GHz, a high frequency geared to small cells and urban build-out, not the long ranges required for rural networks to be cost effective. The problems of delivering data and broadband services to far-flung regions has created an opportunity over the years for mobile satellite providers, but they have their own profitability challenges, because of the high cost of their infrastructure, devices and services. Combining the relatively low costs of wireless networks and standardized handsets with the huge reach of satellite seems like an ideal solution, and one supported by the FCC when it awarded spectrum two years ago to several operators specifically to support hybrid services. One of these is ICO, which is controlled by Craig McCaw, the founder of Clearwire, so a partnership between these two entities - which are already trialling equipment - is certainly a possibility on the US WiMAX horizon. Also bringing WiMAX to this satellite/terrestrial landscape is Globalstar, which had been fighting for the right to support ATC (Advanced Terrestrial Component), the key technology underpinning hybrid services. This was previously only permitted in the special spectrum granted to the ATC operators ICO, MSV and Terrestar, but Globalstar can now support the system in its own frequencies. This, said the operator, will enable it to offer WiMAX to over 500 underserved rural communities in the US, working with its spectrum lessee Open Range Communications. Globalstar and Open Range, as well as having their license terms modified to allow the WiMAX build-out, will also gain a $267m loan from the Department of Agriculture’s Rural Development Utilities Program (contingent on FCC conditions and government approval). “We are very pleased that the FCC has authorized Globalstar and our WiMAX partner to proceed with adding WiMAX service to our mobile satellite service in rural communities throughout America beginning next year,” said Jay Monroe, CEO of Globalstar. “Today’s FCC action is an important step in closing the digital divide in America. We expect our partner to initially deploy infrastructure in more than 500 rural communities with the ability to expand the relationship over the next six years to additional markets covering 50m people or about 15% of the US population.” Globalstar said it plans to pursue discussions with other possible wireless partners in the US and other countries.

Wireless carriers are putting on a brave face even as economic downturn looms, and the US firms in particular are arguing that they can invest their way through recession and come out stronger than before. To do so, though, they are likely to need new networks with advanced performance and low cost of ownership, and this could drive some apparently counter-intuitive upfront investment (which could scare shareholders). Dan Hesse, CEO of Sprint Nextel, is one of the loudest voices behind this argument, and promises no further pullbacks on expansion of WiMAX-based Xohm services (Clearwire in future). The cablecos will be important financial and services partners in the Clearwire effort, and will help steer it through recession, but other operators are taking a similar view, but choosing a different new technology - as Cox’ outline of its ambitious 3G/LTE plans this week indicate. During the last major recession a decade ago, mobile phones were a luxury and people who did access the internet regularly, usually did so via a dial-up modem. Few in developed economies, though, could imagine doing without the fixed telephone line. Now the cellphone and broadband connection have become the essentials, while the fixed line is dispensable. This makes a comparison of the wireless business then and now somewhat futile - but one common factor in all downturns, of course, is that customers will want to cut their spending, while hanging on to what they regard as essentials.

So among telcos, the operator that can combine the two must-haves, the mobile phone and the broadband link, and reduce the user’s overall spend at the same time, will have the best chance of withstanding a market squeeze. This means wireless carriers can target a major wave of fixed line abandonment, but support the massive upsurge in fixed/mobile broadband usage over a wireless connection, an operator must have invested in the most advanced new networks, that can not just support the volumes of true broadband traffic, but also do this at affordable cost of delivery to the operator, and therefore competitive rates for the consumer. WiMAX could well be the first beneficiary, offering as it does a broadband network with low operating costs and low cost per Mbps to deliver services, but still sufficiently powerful to replace fixed lines.

This is certainly the view of Dan Hesse, CEO of Sprint Nextel, which recently went live with its first Xohm WiMAX network in Baltimore, ahead of the merger with Clearwire that will see a national mobile broadband system roll out across the US. Hesse expects this service to cover 140m potential customers by 2010, recession notwithstanding. Hesse told a conference last week: “Now consumers are beginning to cut the cord. In the economic slowdown we are going to see this potentially accelerate. Today, the majority will say the luxury is wireline. Wireline carriers will be impacted more than wireless.” Such viewpoints clearly strike fear into the hearts of wireline-only operators, notably the cablecos, and they need to counter the converged wireless trend by offering affordable fixed/mobile bundles of their own, spreading the load across cable and wireless capacity and supporting the full quad play.

The main US cablecos have all decided that a resale or MVNO deal with a cellco - like the failed Pivot venture the big four had with Sprint - is not enough and they need to have a level of ownership and control of their wireless networks, despite the heavy capex investment needed, and bucking the general worldwide trend for operators to pool spectrum and RAN resources in new markets. So Comcast, Time Warner and Bright House will take stakes in the Clearwire WiMAX-based venture, and look to use this to add mobile broadband and media services from late 2009 under their own brands, increasing customer control through mechanisms like femtocells. But the fourth member of the former Pivot group, Cox Communications, is ploughing its own furrow and looks set to keep pace with Clearwire with a 3G build-out in its AWS spectrum, and plans for LTE in its 700MHz holdings. Cox opted out of the ‘new Clearwire’ and invested heavily in the recent 700MHz auction. It spent about $304m on 14 Block A and eight Block B licenses then, and also invested $248.3m, as part of the SpectrumCo cable consortium, on AWS licenses in an earlier auction. It has now outlined its plans, which involve building a CDMA2000 3G network starting next year, with a view to moving to LTE at the mobile broadband stage, likely around 2011. It will initially roll out CDMA EV-DO, reportedly with equipment from Huawei, in the AWS band, and will keep most of the 700MHz assets for rural coverage and for future LTE.

The CDMA/LTE combination, with the latter focused mainly on data and mobile content services, is becoming a popular strategy for north American operators - notably Verizon and Canada’s Bell and Telus - and this should create some economies in the LTE market, as well as driving early trials (though another LTE adopter, AT&T, does not believe the system will be deployed in volume until 2013). It also creates a north American 4G map that is clearly divided between LTE - Verizon/Alltel, AT&T, Cox, T-Mobile and Bell/Telus - and WiMAX (Sprint, the Clearwire cablecos, Rogers/Inukshuk). Before Cox’ CDMA infrastructure is in place, it will address the immediate demand for cable/wireless bundles more conventionally, via a resale deal with Sprint that kicks off in the first quarter of 2009. “Wireless service will be a key driver to Cox’s future growth,” said president Pat Esser in a statement. “We’ve already invested more than $500m to acquire wireless spectrum and to develop the infrastructure and human resources needed to architect our own advanced wireless service.”

Devices will be critical to mobile broadband operators in differentiating their services, as Sprint’s Xohm unit well knows. Although launched only with laptop options, the first unique product for the WiMAX-based network, Nokia’s N810 Internet Tablet, is now in stock and should be ready to purchase next week. But this will just be the start - on the horizon for 2009, we may see products from the two brands that are currently proving the big marketing hitters for the US cellcos, Apple and Google Android. Hopes for a WiMAX Apple device - initially likely to be a Macbook laptop and/or iPod rather than a full iPhone - are raised by reports that Korea Telecom, the other major operator rolling out WiMAX-based mobile broadband, will soon have Apple products for its network, based on its own 802.16e implementation, WiBro.

KT, which - like most east Asian operators - takes a significant role in the development and design of the devices for its networks, has signed a memorandum of understanding with Apple to work together on mobile broadband enabled products. These would initially be for the current WiBro network, but since this will be upgraded to fully standard wave two Mobile WiMAX, as used by Xohm, as it expands, it should be a small step for Apple to create standards-based gadgets. WiBro would initially be embedded in lightweight MacBook laptops, reinforcing the notion that Apple is looking to push web optimized ‘netbooks’ through new operator channels, as its PC rivals are doing. KT and Apple would sell WiBro MacBooks bundled with broadband services through both companies’ outlets. While this would break the dominance of Windows PCs, PDAs and smartphones on the WiBro network, appealing to a new group of users - especially younger, media-oriented consumers - it also shows KT following the traditional cellco model of subsidizing devices in order to tie in customers and lure subscribers to high end data and broadband plans. By contrast, Xohm has been vocal in supporting the role of WiMAX in breaking that model and enabling operators, through the IP economics of the new standard, to support open access from any user device, and to move away from subsidies. Apple is also looking to put WiMAX into the iPod media player, as it has already done with Wi-Fi, and this could pave the way for a future iPhone. Meanwhile, the other hot phonemaker in the US is currently Taiwan’s HTC, thanks to being the first company to roll out and Android smartphone, the T-Mobile G1. This should prove a major revenue boost for the company, and its first serious move away from Windows devices, the mainstay of its well regarded Touch range of smartphones. Now HTC, insiders say, is planning a Windows Touch device for WiMAX, targeting Xohm and Russia’s Scartel, and aims to produce an Android/WiMAX product next year. From HTC’s point of view, these two moves would be in line with its strategy of stealing a march on larger players by supporting minority technologies at an early stage - it was the first cellphone manufacturer outside the pure ODM community to create Windows handsets, using them to establish a high end range under its own brand, rather than focusing entirely on white label devices, its original business model. But it is still ready to submerge its brand in order to gain early presence in a new and promising market, ahead of the competition, as it has done with G1. Given Xohm’s devotion to open access, a WiMAX Android phone would not only enable HTC to be a first mover again, but also to launch an Android product under its own name. Android products for Xohm are widely expected for mid-2009, given that Google and Sprint Nextel have collaborated closely on the software platform and user experience, and Google is a major supporter of the open access web services model on which Xohm is pinning its hopes. On the Windows front, the WiMAX Touch is reported in various leaks to be called the HTC T8290 and to come with a large 800×480, 3.8-inch touchscreen, a five megapixel camera, built- in GPS and accelerometer, and triple-mode support for WiMAX, Wi-Fi and GSM. If the GSM element is correct, this would clearly be a handset targeted to more operators than Xohm, whose wide area roaming would naturally be to Sprint’s CDMA network (and, with Sprint promising CDMA/WiMAX devices some time next year, surely HTC would create a CDMA version too?) Indeed, the first customer for the HTC product is slated to be Russia’s Scartel, which is rolling out WiMAX in Moscow and St Petersburg under the Yota brand.

Only two weeks ago Nortel was reported to be poised to get out of wireless networking, but the Canadian firm does not throw in the towel easily, and managed to make a splash at last week’s WiMAX World show even though it no longer plans to make its own WiMAX products.

The company has always preached the importance of building a broad ecosystem around WiMAX at an early stage, in order to differentiate it from other networks through a well differentiated choice of devices and applications. Now it is stretching the ecosystem concept so far as to rely on a third party - Alvarion - even for the RAN infrastructure itself, but still insists it can play a pivotal role as integrator and as the hub for an innovative line-up of partners.

The Nortel WiMAX ecosystem was expanded last week to include such diverse partners as IBM, core networking specialist WiChorus, Quanta Computer and Accton Wireless Broadband.

Between them, and with the addition of a group of device makers, Nortel says it will offer an end-to-end platform, and that this best of breed approach - with the key supplier not actually providing the hardware - represents a blueprint for next generation networks, that many will emulate. Logically, this would mean that infrastructure would become increasingly commoditized and the preserve of a few vendors, either with the scale to thrive in the hardware game, or with specialist expertise in an emerging air interface technology - and vendors would gain value and differentiation from providing software, services and integration. This is a trend that has already been clearly seen in the enterprise computing and networking fields, with even hardware majors like IBM now primarily focused on services.

Nortel has OEMd Airspan gear for fixed WiMAX for some years and earlier this year abandoned its own Mobile WiMAX development plans for a similar agreement with Alvarion for 802.16e. This was widely seen as a downgrading of WiMAX’ importance in the Nortel strategy, which had initially rested on stealing a march in 4G - having exited W-CDMA - by creating a unified OFDMA platform for WiMAX and LTE. LTE then moved to the center of the Nortel roadmap, though even this was put in doubt last month when CEO Mike Zafirovski announced his intention to sell off various units, possibly including wireless, and focus on areas where Nortel can still lead the market, like enterprise unified communications and carrier VoIP.

“For Nortel, our sweet spot is the integration of a complete solution,” said Scott Wickware, WiMAX general manager at Nortel.  “Backhaul, applications like VoIP, professional services, devices and enterprise offerings are all part of a WiMAX build and where Nortel excels.”

One of the most strategic areas of Nortel’s portfolio is unified communications, and particularly its alliances with IBM and Microsoft. Now it is extending its UC strategy to new networks, including WiMAX, and is working with IBM to deliver Lotus Notes capabilities over 802.16e. This will support a range of services including IP telephony, instant messaging, web conferencing, video chat and unified messaging, regardless of location and device. 

Other new partners include WiChorus, for its Home Agent, which provides operators with subscriber and content management. On the device front - where Nortel was an early mover in terms of attracting Taiwanese ODMs into the WiMAX fold to boost the economics - the vendor is working with Quanta and Accton in Taiwan to deliver Nortel-branded 802.16e devices including PCMCIA cards, USB adapters, indoor gateways, window mount antennas and outdoor gateways working in the 2.3GHz, 2.5GHz and 3.5GHz bands.

It remains unclear whether Nortel will stay in the wireless infrastructure market long term, or seek to expand its partnership with IBM and become a wireless integrator for enterprises and operators. So far, it insists it still plans its own LTE products, and Danny Locklear, VP of marketing for carrier networks, said LTE gear is already being trialled with Verizon Wireless and T-Mobile Germany. “We see a huge global opportunity with LTE,” he said. “WiMAX is a good opportunity as well, although not as large.” However, he said in an interview that there could be several ways forward with LTE, including a partnership similar to those with Alvarion and Airspan. There will be fewer choices in LTE of a hardware partner sufficiently small for Nortel to hold the upper hand - LTE is likely to be the preserve of the big five manufacturers from an early stage and they will have their own eyes on building an integration business. The Canadian company is likely to look to Asia for its alliances - it is hinting at a “go-to-market” deal with an Asian network firm, and its LG Nortel joint venture is already working on LTE devices. Whatever its decisions on making infrastructure, Nortel knows that, to hold a major position in LTE, it will have to offer a strong choice of devices, avoiding the mistake it made in WiMAX, where early adopters like Sprint were looking only at suppliers that could offer terminals as well as networks.

Speculation that Nokia Siemens Networks (NSN) will defocus on the WiMAX market has gained new weight, with Sprint Nextel defecting to Samsung to supply the Dallas-Fort Worth build-out for its Xohm service. The operator insists this is just a practicality, because the Texas city became ready for its equipment ahead of schedule, and there was existing inventory of the Samsung gear, which has been used in the first Xohm territory to go live, Baltimore. Nearly two years ago, Sprint chose three vendors to supply the infrastructure for Xohm, and each gained an initial city to build out and put WiMAX systems through their paces. Motorola won Chicago, Samsung Baltimore/Washington DC, and the surprise third supplier was NSN (then still Nokia), which at the time had WiMAX kit only on paper, and was far behind other candidates like Nortel in developing the technology. However, Sprint clearly wanted to sign up companies that could also make innovative devices, such as the Nokia N810 Internet Tablet for WiMAX, and NSN boasted that its flagship 3G range, the Flexi Base Station, could be easily adapted for WiMAX and for low-cost roll-out into the bargain.

Despite the clear competitive advantages of the compact, modular and multi-network Flexi - which could be a breakthrough product for NSN in the usually problematic US, with T-Mobile USA choosing it for 3G - NSN’s commitment to WiMAX remained under question, especially as Nokia is such a fervent champion of, and likely beneficiary from, LTE. It would surprise very few people if the Texas change is the first step in a quiet exit from Mobile WiMAX for NSN, especially as its Finnish parent will still be able to pursue the primary agenda, gaining increased US market share for its handsets. Participation in Xohm was seen as a possible weapon in this fight, but in reality, as Verizon and AT&T move towards LTE, Nokia is likely to see those operators, and the direct channel, as the most important vehicles for pushing its devices across the Atlantic. In a research note when Nokia won the Dallas deal, analysts at Dresdner Kleinwort wrote: “Nokia, we infer, views the WiMAX experiment mainly as a vehicle to regain handset share. A recovery of the North American mobile device operations has long stood, and continues to stand, on the top of Nokia’s strategic agenda. Whether Sprint Nextel constitutes the right vehicle to effect a positive change is more debatable [given low volumes in the early years].” All NSN would officially say about the Xohm decision in Dallas was that it understood the reasons for the change, given that the city’s launch was ahead of schedule (though no date has been given). “Xohm’s Dallas market was ready for deployment ahead of schedule, so it made sense that they would use existing inventory to accelerate the deployment,” said Chantal Boeckman, NSN communications manager. She insisted that Flexi is performing well in Xohm trials and has embarked on the certification process. “We continue to be an active member of the WiMAX ecosystem,” she said. “We’re confident Nokia Siemens will be part of the new Clearwire deployments moving forward.” This week sees the official launch of Xohm in Baltimore, though some users gained early access to the service a week ago. There was real excitement surrounding a relatively small launch that could, nonetheless, be the starting point for an important roll-out, in terms of bringing true mobile broadband and open access to the US, and shaking up the traditional telecoms landscape, so dominated by AT&T and Verizon. There is even talk of Android devices making it on to Xohm as early as mid-2009, which would be logical, given Google’s work with Sprint on the user experience for the network, and its planned investment in the new Clearwire. A WiMAX Android device would, however, carry symbolic and marketing weight, emphasizing that Xohm aims to accelerate, and grab the reins of, the move towards a fully open mobile internet, where any terminal and application can run on the network, and a wholesale model can support small and innovative services. Nonetheless, Baltimore is a game changer with very modest beginnings. It is offering DSL-class fixed or mobile broadband services on a month-by-month, not contract basis, with speeds and prices that are comparable to many of the emerging HSPA or EV-DO offerings from other cellcos. It is available in Maryland’s largest city and the surrounding county, via, initially, just two devices, a $60 Samsung laptop aircard and an $80 ZyXel USB modem. More famous early WiMAX devices, the Nokia internet tablet and the Intel WiMAX/Wi-Fi Centrino for embedded laptop wireless, should follow soon once they are through WiMAX Forum, FCC and carrier certification. Other Xohm markets will follow later this year and early in 2009, led by Boston, Chicago, Dallas Fort-Worth, Providence, Philadelphia and Washington DC. Service plans for Xohm, which do not require a long term contract, include a $10 day pass, a $25 monthly home internet service and a $30 monthly mobile service, with rates of 2Mbps-4Mbps. More details of the devices and services plans will be announced at the official launch. Sprint has been an early cheerleader for the fundamentals of the mobile web vision - open access, flat rate tariffs, unfettered roaming, freedom to go beyond a carrier’s portal. But so have many other companies, large and small - the difference is that Sprint now has a network on which to demonstrate that this vision actually works in reality, and is actually usable and attractive to consumers and businesses.

Sprint and Clearwire may grab most of the US WiMAX headlines, but there is growing uptake of the technology among smaller operators using the new lightly licensed 3.65GHz band, and this may support a more attractive business model than the last system for delivering disruptive open wireless services, metro Wi-Fi mesh.

Last week, a small municipal provider, Azulstar, announced that it has already rolled out WiMAX-based services in various towns in western Michigan and New Mexico, including its home base of Grand Haven, Michigan. Like some other early adopters of the 3.65GHz band, Azulstar has grasped the opportunity to enhance a model that originally depended on Wi-Fi - or more significantly, on unlicensed spectrum in 2.4GHz or 5GHz - by using a less crowded and higher quality band, and a more ‘grown-up’ technology.

The commercial aims will be similar to those of many Wi-Fi mesh users - to use a relatively low cost, all-IP technology to deliver affordable, widely available internet services, with various levels of service from basic cheap or free access, through VoIP and multimedia, up to business services and even mobile hand-off within the carrier’s zone or roaming partnerships. This would provide a viable, and differentiated, alternative to carrier services, especially for underserved communities or small/medium enterprises looking for a more cost effective alternative to telco offerings.

All this appeared to make sense in Wi-Fi, and for a while telcos seemed deeply concerned about the competition, and price depressing tendencies, of open access alternatives. But despite some Wi-Fi success stories, often tied into public safety requirements, the use of license-free spectrum always carries the risks of interference and severe limitations on ensuring quality of service - plus virtually no barrier to entry for other competitors, leading to a constant price cutting spiral that also leads to deterioration of quality and to increased congestion.

Enter the 3.65GHz option, peculiar to the US - which, unlike most of the world, does not have the 3.5GHz band open as a licensed broadband wireless frequency. The 3.65GHz spectrum comes with a light license - ‘good neighbor’ obligations that are sufficient to prevent interference and to allow providers to sign service level agreements and ensure different levels of QoS for different customers. Yet it is very low cost spectrum, and sufficiently close to 3.5GHz that vendors say they are easily able to adapt their kit for that band - the most widely used in the world for WiMAX, and likely to remain that way for many years until 2.5GHz is in broad use. This means that, even though 3.65GHz is a single-country frequency, it will be able to take advantage of the economies of scale of the 3.5GHz WiMAX kit, reducing costs for carriers.

Azulstar, for instance, is being supplied by Alvarion, with its 802.16e BreezeMAX 3650 base stations and CPE, a variation on its existing BreezeMAX equipment for 3.5GHz. Other vendors that tend to target smaller carriers are also turning out systems, including Aperto. The provider has been testing its new service since February in the West Michigan area, and it is now formally launching in the communities of Grand Rapids, Grand Haven, Ferrysburg and Spring Lake, among other locations. In New Mexico, it is launching services in Rio Rancho and Albuquerque, and it has additional markets under construction.

These services will focus on internet and voice for businesses, homes and local government customers, with prices starting at under $275 for businesses, and phone and internet home bundles beginning at $40 per month. It promises symmetric data speeds of 100Mbps for business and speeds of up to 6Mbps for domestic users.

Some of the companies acquiring 3.65GHz licenses - now that we have seen the end of a lengthy debate over anti-inteference rules for the band, which threatened to exclude WiMAX at one time and delayed uptake - are traditional broadband wireless access providers, moving away from proprietary kit; some are new entrants or small wireline carriers; but many are trying to improve on a model that previously relied on Wi-Fi. Azulstar is one of these. It was an early mover in municipal Wi-Fi, having launched services in Grand Haven five years ago, but says - echoing many similar local operators - that WiMAX will offer better reliability, a wider range of grades of performance, and the ability to target specific user groups more easily. It will also, once 3.65GHz 802.16e gains a measure of volume, work out as cheap as Wi-Fi mesh to deploy and run - or even cheaper, especially when considering five-year cost of ownership, and systems that need to support VoIP and video as well as basic data. Azulstar was also early into WiMAX, working closely with Intel - for instance, in 2004 it rescued a project to roll out a metrozone in Rio Rancho, New Mexico, where Intel has several facilities, from the original contractor, and said at that point that it would improve the economics by using a hybrid WiMAX/Wi-Fi network.

Azulstar now plans to upgrade all its existing municipal Wi-Fi networks with WiMAX equipment from Alvarion, Airspan and Redline, rolling out new networks in 3.6GHz across 15 cities in the Midwest and Southwest. In these areas, it will overlay the existing Wi-Fi mesh, which will continue to operate, supporting visitors and residential customers as long as there is demand. Legacy Wi-Fi customers wanting to move to WiMAX will receive a discount toward their adapter. Azulstar now says it will only own and operate municipal systems where it can get spectrum and deploy WiMAX, and in markets where it cannot access 3.6GHz (or other future low cost licensed spectrum options) it will just take on a subcontractor role, using Wi-Fi but with no network ownership. 

The company may prove a blueprint for many local providers, and possibly for a future roaming partnership across large swathes of the US, especially in rural areas. Clearwire itself became involved in the municipal market when it won a contract to build out Grand Rapids, also in Michigan, in preference to a Wi-Fi approach - with the primary decider being the ability to offer many levels of services. Other early users of 3.65GHz spectrum include NextPhase Wireless in California, which has gone live in Orange County, plus Towerstream and Pipeline Wireless. Towerstream is an example of a provider that has made a success of fixed WiMAX in the unlicensed 5GHz band, but wants to move to the volumes and mobile potential of 802.16e, which will not be widely available in that frequency.

The first round of 3.5GHz spectrum auctions in western Europe, mainly around the start of the decade, was largely a failure, with too many licenses and inflated prices being allocated in the wake of the telecoms boom. The emergence of a standards-based technology, WiMAX, for the band, reignited interest in 3.5GHz, with the potential to add nomadic/mobile services and reduce cost bases, all combined with the rising demand for universal broadband access. But while many countries have held new rounds of auctions - and in countries like Germany, Holland and the UK some advanced trials and even commercial services are rolling out - others have not avoided the mistakes of the past, with France the most glaring example.

Until late 2006, France had a single national 3.5GHz license holder, Altitude Telecom, which has made strong progress with its roll-out since becoming part of the Iliad broadband group, itself integrating WiMAX into its disruptive quad play plans. But then more licenses were auctioned, creating a total of 19 regional franchises holding 44 individual licenses, which may have fragmented the market beyond the point of commercial viability, especially outside rural areas underserved by broadband, or enterprise hotzones.

Two years after the auctions, regulator ARCEP has found that only three of the 19 license holders have met their WiMAX roll-out obligations, with only about 15% of the promised sites operational by the end of June 2008. According to studies by ARCEP and UK-based analysts at Ovum, the most aggressive buyer of 3.5GHz licenses, Bolloré  Telecom, had deployed just 11 sites by the end of June instead of the 968 sites it had targeted.

Meanwhile, nine spectrum owners have done nothing at all and, of the 3,564 sites that were supposed to be operational by the end of June, only 526 were up and running. The market is now far too fragmented, with some smaller operators looking non-viable for the costs and risk involved in rolling out new services in a country that is already well served by fixed and mobile broadband and by enterprise offerings. One factor, according to Ovum, has been that regional public authorities that won spectrum in 2006 have been distributing their rights at a department level, upping the number of players from 10 to 19, despite some consolidation (notably Bolloré buying eight licenses from HDRR/TDF earlier this year).

ARCEP said it will continue to monitor WiMAX roll-outs until the next deployment obligation deadline, scheduled for December 2010. In June it said it would investigate the progress of operators to see whether they were meeting their obligations - and those failing the test could then face suspension of the license, or a fine equal to 3% of revenue. However, it now seems that ARCEP will give the license holders another two years to get their acts together by building out or selling on their franchises, and we could see a wave of consolidation and, then, a better chance of national or major regional WiMAX services.

Some operators, despite the slowness of real commercial progress, have already been adapting their business models to rise to the challenge more effectively. In the summer, the second largest regional license holder, SHD (Société Haut Débit) - jointly owned by cellco SFR and telco Neuf Cegetel - announced a virtual network operator agreement with the regional broadband group Numéo, covering the Ile-de-France region around Paris and the center, and the Provence-Alpes-Cote d’Azur region in the south. This will allow Numéo to bundle WiMAX services in regions where it cannot provide ADSL. The company has become a significant provider of rural internet access over the past four years, and is well positioned to become the largest provider of rural WiMAX services by early 2009.  It already has similar agreements with the other main WiMAX spectrum owners, HDRR and Altitude/Iliad. These cover the cities of Loiret, Seine-et-Marne and Quimper, and an additional eight regions are expected to follow. 

A deal with Bolloré -  90% owned by Group Bolloré and 10% by Hub Telecom, a subsidiary of Aeroports de Paris - would create an almost-national system and the opportunity to support wholesale and enterprise plans. Bolloré now has 20 licenses, having acquired eight of the 10 initially snapped up by HDRR. It blames lack of certified terminal equipment for the delays in its plans to support national broadband internet services for businesses and consumers.

Ovum commented: “At the time of the license allocation in July 2006, we commented that we had difficulty understanding ARCEP’s vision regarding WiMAX, and its expected impact on the French telecoms landscape. This was because most of the licenses were allocated to new entrants rather than established French or foreign telecoms players.

Furthermore, the geographical segmentation of the spectrum added complexity, with none of the selected companies able to provide nationwide coverage. There has since been some consolidation, with Bolloré Telecom purchasing eight regional licenses from HDRR to achieve almost nationwide coverage (two regions are still missing).”

One of Vodafone’s key growth strategies is to move into fixed line broadband and fixed/mobile convergence. While it has been buying or partnering with fixed operators in many of its key territories, it can make faster progress in new areas where it can acquire providers more cheaply and has a more open market to play for. So as part of its Middle Eastern push, it has bought a fixed line license in Qatar to add to the mobile franchise it gained last December, in a ploy that is likely to be a blueprint for many countries in the EMEA region. What’s more, the license includes a mandate to provide WiMAX, at least for fixed data services, in a bid to increase broadband penetration rapidly.

While Vodafone has not shown much interest in WiMAX in its heartlands, it has worked with the technology for fixed/mobile convergence trials and in some newer territories, including through its Middle East joint venture with Zain, based in Bahrain. In Qatar, the regulator, ictQatar, has decided to package WiMAX with the new fixed line license to give the new entrant a differentiation over incumbent Qatar Telecom. Also, according to the agency’s executive director William Fagan,  the decision was taken because “Qatar is too small to have a separate WiMAX licensee … or for a WiMAX operator to compete with two fixed line operators.”

He added, in an interview with Gulf Times: “We thought if the second operator can combine the WiMAX with the fiber that is going to be put into Lusail and The Pearl Qatar among other places, we can complement the two services.” This is a combined approach that is set to be increasingly common in developing markets, where the flexibility for operators to use broadband wireless technologies as well as copper or fiber can significantly speed up the roll-out of high speed internet services, especially in rural or sparsely populated areas, or difficult terrain. The most dramatic example is India, where state-owned BSNL and other operators are looking to use WiMAX to meet their aggressive broadband build-out targets, as well as gaining the ability to support nomadic and mobile services in future to increase ARPU compared to fixed-only providers.

Vodafone won its new license - like the mobile one - in partnership with the Qatar Foundation Consortium. In both cases, it is only the second licensee, breaking the former monopoly of Qatar Telecom (Qtel), in this small but rich and influential state. Qtel is itself eyeing an international expansion program in the Middle East, Africa and Asia and recently acquired Indosat in Indonesia. Last week it reported a 102% increase in first half group revenues, to QAR8.1bn ($2.2bn), with net profit up 33% year-on-year at QAR1.18bn  ($324m).

Financial terms of the Vodafone deal were not disclosed, although a Reuters report states that Qatar’s telecoms regulator, the Supreme Council of Information and Communication Technology, said in December the fixed license would be sold for a fixed fee of QAR10m ($2.8m) and services should start in 2009.

The mobile joint venture with the Foundation will start GSM services in the first quarter of next year. Vodafone owns 51% but the venture is set to sell 40% of its shares to the public before the end of the year.