Published: 4 May, 2011
By virtue of being one of the first and largest mobile WiMAX operators, Clearwire has always been seen as the key bellwether for the technology. Then, its decision to trial the other 4G option for TDD spectrum, TD-LTE, helped endorse a trend for WiMAX players to examine a multi-technology future. The problem for the sector has always been that Clearwire is so atypical, with its huge spectrum assets and tier one shareholders, and its dual (and often conflicting) role as Sprint’s 4G network and the pioneer of a new, open broadband model in the US. The future of Clearwire’s network will only become clear when Sprint makes its decisions about its own 4G future - and whether to acquire its joint venture outright, dump it, or stick with the status quo. In the meantime, the rest of the WiMAX players need to make their own technology decisions in the coming months, and may find smaller operators a more realistic benchmark to use.
It is clear that, for new deployments by mobile operators in TDD spectrum, TD-LTE will be the natural option. It will give greater synergies with any FD-LTE deployments or partnerships they have, and potentially will tap into a converged LTE ecosystem (though China Mobile’s pleas to the device industry to embed both spectrum modes into every device from day one are falling on deaf ears).
However, there are two myths about the WiMAX/TD-LTE stand-off. One is that current WiMAX operators will ditch their technologies rapidly to embrace TD-LTE alone. In fact, coexistence will be far more important than an either/or approach for successful vendors in this field. This was seen when Ericsson and Alcatel-Lucent were forced to license Samsung’s WiMAX technology as part of their role in Sprint’s huge multi-technology Network Vision program. And Huawei has been particularly quick to see the potential for a dual-mode approach, offering WiMAX alongside 3G and both LTE flavors in its flexible base station architecture. This has been rewarded with the Clearwire trials, clearly looking to phasing TD-LTE in alongside WiMAX rather than stripping out the latter any time soon, and with other deals such as one with Vividwireless in Australia.
In many ways this will be a better case study for TDD spectrum holders than Clearwire, since it demonstrates what will be a very common and pragmatic approach among those who have invested in WiMAX for mobile broadband (as opposed to fixed-only deals, where WiMAX will remain the key platform as a DSL alternative). Vividwireless, formerly known as Unwired Australia and now part of the Seven media group, has completed its WiMAX build-out, addressing the central zones of the six main Australian cities (it recently completed Adelaide, Canberra and Brisbane, and already had activities in Sydney, Melbourne and Perth). It is trialling TD-LTE with a firm eye on coexistence, looking to preserve and enhance the WiMAX metrozones and then use TD-LTE to expand coverage citywide and to reach into the suburbs. A single technology platform may be the final end game, but that will come in many years’ time.
Also chasing the coexistence theme are WiMAX infrastructure vendors such as Samsung, Alvarion and Motorola, all of which have TD-LTE products, though not fully converged software base stations like Huawei. Closer to that goal is ZTE, which recently demonstrated WiMAX/TD-LTE convergence on one network. It was working with Malaysian WiMAX operator Packet One Networks, which claimed to be able to upgrade its current infrastructure to support TD-LTE with no hardware replacement at all.
"WiMAX has enabled P1 to become a 4G leader in south east Asia. Our next step is to become the first company leveraging the power of both WiMAX and LTE,” said CEO Michael Lai. Like other carriers, P1 knows flexibility is crucial to ensure return on the first round of investment and the best timing for future ROI. Lai continued: "The decision to deploy LTE will depend on other major operators in Asia. In other words, it will depend on the maturity of the ecosystem when operators around the world, such as China and India, begin the commercial application of TD-LTE as well as also the availability of LTE devices."
The second myth is that, bar some capacity upgrades here and there, WiMAX has no market any longer outside fixed connections for underserved communities. In fact, the maturity of the technology, and the existence of an open and cost effective silicon ecosystem, makes it well suited to provide readymade platforms for high growth markets where availability of mass volume consumer devices are less vital. The smart grid has already proven a rich seam to mine for WiMAX, as some utilities look for dedicated and reliable networks that are independent of the cellcos. And there is also the prospect of a second coming of the municipal broadband wave, incorporating WiMAX as well as the more traditional Wi-Fi (and at lower cost than fiber).
Some of Clearwire’s less publicized activities are in this area, and another example is seen in Houston, Texas, where a 640-square mile WiMAX network is to replace a previous Wi-Fi mesh, built in 2007. The new program will cost $6.4m, mainly funded by broadband stimulus grants, and will provide full coverage for the city plus a replacement for leased lines for the local authority’s networks – running applications like meter reading and traffic management. “I think the payback time is easily within 24 months,” said program manager Brian Anderson, who is using Alvarion BreezeMAX kit in the lightly licensed 3.65GHz band. The previous Wi-Fi metrozone had failed when its operator, EarthLink, pulled out of the sector. It had found the capital expense of installing Wi-Fi for true universal coverage, not just hotzones, was prohibitive and paid Houston $5m to extricate itself from what had been the flagship in an ambitious city WLan strategy.
So WiMAX will pursue its own markets, while the mobile broadband players weigh up their options. And though Ericsson CFO Jan Frykhammar recently proclaimed 2011 to be “the year of TD-LTE”, the technology remains immature and the period up to 2014, at least, is more likely to be characterized by dual-mode projects. Even Frykhammar said he did not expect commercial TD-LTE networks this year.
The biggest influence on how quickly the TD-LTE ecosystem evolves will be flagwaver China Mobile, but even that giant has become rather more cautious about its timescales recently. The other key territory will be India, once expected to be a WiMAX domain but now firmly focused on TD-LTE.
The largest privately held owner of India’s BWA spectrum in 2.3GHz, Reliance Infocomm, looks set to rely mainly on TD-LTE, and even the plans of the state-owned operators - BSNL and MTNL - which have already started WiMAX build-out to address broadband shortages, are in some disarray. Amid government proposals to merge the two carriers with one another and state manufacturer ITL, MTNL is reported to have postponed its WiMAX build-outs in Delhi and Mumbai indefinitely (it holds 2.6GHz spectrum in these two huge metros, while BSNL has the licenses everywhere else). “We have decided not to go ahead with our WiMAX plans for the time being," a senior MTNL executive told Light Reading.
BSNL has missed its March deadline for launching commercial urban WiMAX services (it does have a separate WiMAX program running in rural parts). But it has appointed franchise partners and looks set to continue with its plans.
Many TD-LTE build-outs will be focused mainly on portable or fixed broadband, and can thrive with modems and dongles. But to appeal to Clearwire, or to some of its early cellco triallists such as Hi3G in Sweden or of course China Mobile, handsets will be a must. Yet, despite the achievement of getting strong smartphones onto Verizon’s LTE network at such an early stage, even in FD-LTE the device ecosystem will take time to develop.
There will be 16m LTE subscribers in the world by the end of this year, according to forecasts from ABI Research, and about 100 LTE-ready devices are in the market. But only 13% of those are smartphones or tablets, which will be key to driving mass uptake. In 2014, ABI thinks, 205m LTE mobile devices will ship, 72% of them being handsets and tablets. Sticking to those kind of timescales for mass market LTE smartphones will depend on the chipmakers.
For now, they agree they are in the early stages of making LTE chipsets really affordable and fully functional. Basebands are not yet tightly integrated, as in 3G, and Apple’s acting CEO, Tim Cook, recently justified the lack of a 4G iPhone for Verizon by saying that would involve too many "compromises in design”. Will Strauss, principal at chip analyst firm Forward Concepts, has been analyzing the silicon players poised for when the 4G handset space takes off. There are 11 with LTE basebands, or integrated baseband/processors, some having entered the market by snapping up pioneers in WiMAX modems (Broadcom bought Beceem, and Cavium acquired WaveSat). While Qualcomm and Samsung are the predictable leaders, the rest of the list includes Intel, courtesy of its Infineon Wireless buy; Renesas Mobile, which now owns Nokia’s former modem operation; Japanese carrier NTT DoCoMo (via its silicon partnership with Fujitsu, Panasonic and Renesas); plus MediaTek, ST-Ericsson and two other WiMAX specialists, Altair and Sequans. Not on the list, but a frontrunner in LTE modems, is UK-based Icera.
Strauss echoes the sentiments of Apple’s Cook and endorses the view that a 4G iPhone should wait for a more mature market. He believes the Qualcomm and Samsung chips that are shipping in Verizon’s first handsets are “compromised”. For instance, Qualcomm's MDM9600 is used in HTC's Thunderbolt handset but is not integrated with the Snapdragon processor/CDMA modem. This impacts battery life as well as cost and size.
The headstart that WiMAX players gained in OFDMA technologies mean that, behind the two giants, these specialists are next in line. Sequans, Altair and Cavium are all sampling their LTE chips, as are NTT DoCoMo and Renesas. According to Strauss, STE will come next, during the current quarter, while Intel will sample in the third quarter, and Broadcom and MediaTek early in 2012. Icera began sampling in January.
Motorola Solutions celebrates independence with 11n upgrade
Motorola Solutions gave an encouraging account of itself in its first quarter as a standalone company, reporting an 8% rise in sales to $1.9bn. Having sold off most of its wireless networks activities to Nokia Siemens, a transaction that finally closed last week, the company expects solid growth in its key remaining businesses, government solutions and enterprise Wi-Fi – though iDEN, which it retains, will decline. Operating profit was up to $170m, from $120m in the year-ago quarter.
The company also upgraded its corporate WLan range, boasting a tenfold increase in capacity for its 802.11n controller the NX 9000. This can now handle up to 10,000 access points and the refresh of the range is designed to make large scale WLans more affordable and reliable. There are new single- and dual-radio 11n access points, and the products incorporate Motorola’s WiNG 5 software, which aims to push more intelligence to the edge of the network, the latest trend in large WLan design.
The new NX 9000 Integrated Services Controller significantly boosts the capacity of its predecessor, the RFS 7000, which handled about 1,000 APs. The new box handles AP configuration, security network management and other facilities.
Meanwhile, NSN has wasted no time in providing a migration path for its new Motorola Networks client base. It has launched a product to connect existing Motorola GSM base stations to its Flexi controller. This will boost performance in current 2G networks and give them a software defined path towards 3G or 4G via NSN’s Flexi programmable, multi-network base stations. The solution will be commercially available at the end of June. The company says one Flexi Base Station Controller is the performance equivalent of 30 Motorola controllers, reducing operational cost and improving user numbers and network quality.