Two months after the 802.16e mobile WiMAX standard was frozen, it was officially approved by the IEEE last week, a welcome milestone that steps up the pressure on the WiMAX Forum to ensure the platform becomes commercially viable as quickly as possible. This will mean incorporating 802.16e into the testing and certification schedule rapidly, and making some key - and difficult - decisions on how the new standard is positioned alongside the existing WiMAX standard for fixed wireless access, 802.16-2004.

The fact that the two WiMAX standards are now only distantly related, since the Forum adopted a different physical layer (Scalable OFDMA) for the mobile version, has caused many complications in the roadmap, which was once intended to offer a smooth transition path from fixed to mobile systems. There are real markets for 802.16-2004 -among traditional broadband wireless access (BWA) providers moving from less efficient proprietary technologies, and WISPs in the 5GHz band, as well as in cellular and enterprise backhaul, so the older specification cannot be written off altogether, especially as traditional BWA vendors need a short term revenue stream and so are pushing it hard.

But the main interest from major operators and suppliers is in the more disruptive 802.16e, which will, in future iterations, support full mobility and also deliver all the benefits of 802.16-2004 in a fixed and portable environment. That leaves two issues - what will the WiMAX community do, if anything, for service providers that have invested in fixed WiMAX and risk being stuck with a dead end technology once their vendors shift their focus to 'e'? And how long will those operators that want to move straight to 'e' have to wait for realistic products?

There are some efforts to ensure that 802.16-2004 does not languish, after 2006, into being a niche WISP technology supported by just a few vendors and therefore losing the volume economics associated with open standards. The main hopes for early investors in 802.16-2004 rest with the importance of the laptop agenda, in which the older standard has an important role to play; and vendor development of techniques that will enable fixed and mobile subscriber units to work with the same base stations, even if there is no direct upgrade path. If an operator can be sure that its subscriber units will not have to be replaced all at once if it moves to a mobile base station, the economic pain of replacing the infrastructure will be lessened, as well as the impact on customer satisfaction.

Backwards compatibility:

Supporters of 802.16-2004 believe that there will be early stage demand for laptop-based WiMAX, based around metro area hotspots on a similar model to Wi-Fi, and that this will ensure a healthy market for the fixed standard, since it will be present in PCMCIA cards before 802.16e gets there. Chipmakers like Wavesat and TeleCIS, which have pinned many hopes on 802.16-2004 developments, will get laptop and other embedded devices to market by mid-2006, using techniques such as sub-channelization to improve the portable capabilities of the older platform and giving it a realistic life expectancy before 802.16e takes over.

Preserving 802.16-2004 life through subchannelization

Even a two-year lifespan will not reassure all service providers, who face the dilemma of whether to invest now and get early mover advantage, but risk being rapidly lumbered with a legacy technology, or wait until 802.16e systems are economical and effective before investing at all. Only the large providers really have the option of buying mobile WiMAX in the earliest days, when prices will be high, and the other possibility, investing in a proprietary mobile broadband technology such as Navini and then trusting to its WiMAX migration path, can also carry risks of timescales and upgrade costs that may be unacceptable for smaller operators.

Evolutionary Taskgroup:

Therefore, some assurance that there will be backwards compatibility is essential and Wavesat is a prime mover behind a new initiative, the 802.16e OFDM Evolutionary Taskgroup, which is addressing this issue in order to raise confidence in the first WiMAX platform and stimulate sales. The aim is to encourage the use of 802.16-2004 with subchannelization for the move from fixed to portable operations, and then support a gradual shift to 802.16e that will allow old subscriber units to work with 'e' base stations, in portable mode only, to avoid a mass replacement.

Wavesat's chief scientist, Dr Jonathan Labs, has been named chair of the Evolutionary Taskgroup (ETG), which will focus on developing 802.16e profiles providing compatibility for operators' 802.16-2004 deployments.

"An evolutionary path from fixed to portability will boost the Wi-MAX momentum and be much more appealing to operators, providing continuity and the ability to upgrade service by protecting their investment," said Labs. "The key is to develop an 802.16e OFDM profile backward compatible with the 802.16-2004 standard. In essence, we want to ensure that new base stations will be able to talk to earlier versions of subscriber stations while providing newly added capabilities to new subscribers. The same way, new subscriber stations will be able to communicate with earlier version base stations."

The ETG is part of the WiMAX Forum's sub-11GHz Technical Working Group, which also includes the Mobility Taskgroup.

SR Telecom:

The need for traditional BWA vendors and operators to keep 802.16-2004 alive is indicated by the decision of SR Telecom to launch a fixed WiMAX product, having originally said it would only support WiMAX in its symmetry pre-standard range at the mobile stage. Although marketing chief Chad Pralle still believes that the 'e' variant is superior, even for fixed access, to its predecessor, he said smaller vendors need to move early to gain a "beach head", especially among the larger carriers that SR mainly targets. Its key customers include Telefonica in Spain and some parts of Latin America, and Telmex in Argentina, and these are the sort of large operators that would naturally turn to the major suppliers once those have products in the market. Therefore it is important for smaller vendors to be able to offer them a system in advance of the majors entering the fray - and to provide a great deal of added value on top of the base standards, which in SR's case has focused in particular on VoIP support, redundancy and diversity.

Delays:

The process of ratifying mobile WiMAX has been far more complex and lengthy than originally anticipated, though testing is now expected to start around midyear. The IEEE 802.16 Taskgroup, whose work is now over, studied 12 official drafts, and considered 6,000 comments and 900 documents since it was formed in December 2002 after the original specification was approved.

___________________________

SOFDMA

The key difference between the fixed and mobile WiMAX standards is the more efficient SOFDMA modulation scheme. SOFDMA (Scalable OFDM Access) can assign a subset of subcarriers to individual users. By using different subcarriers multiple people can connect at the same time on the same frequency without interference. The number of subcarriers can adjust dynamically for different conditions.

The Korean Wi-Bro standard was largely developed around SOFDMA and the IEEE became convinced that the advantages of scalable carriers and subchannels were so compelling they adopted it despite the fact that it broke compatiblity with the original 802.16-2004 standard.

Adding subchannelization, and the two key smart antenna methods favored by the WiMAX Forum - MIMO and AAS or beam forming - could increase coverage from two to nine kilometres radius for an urban base station with mobile support, a 20-fold increase in subscriber capacity. The WiMAX Forum has not announced the key spectrum profiles for 802.16e yet but the first are likely to be 2.5GHz and 2.3GHz because these are where the largest planned networks - those of Sprint Nextel in the US and Korea Telecom and SKT in Korea, will be.

___________________________

"We projected completion in about 18 months," said Roger Marks, chairman of the Taskgroup. "Actual development took about twice that long. Though we all would have preferred to stick closer to schedule, the reality is that the working group changed dramatically during those years" - partly, of course, because of the U-turn to adopt Samsung's Wi-Bro technology with its SOFDMA base as a core component of 802.16e. In the three years, the group grew from 82 participants to 310 as interest in mobile WiMAX soared - along with the politics involved - and the variety of applications to consider mushroomed.

Contrast of initial 802.16e and with the addition of subchannelization and smart antennas

With so many interested parties, it is vital that that the process of enhancing the standard and certifying it does not fall prey to similar delays, and that it proceeds more smoothly and with clearer communication to the market than the fixed WiMAX certification has. Improvements to 802.16e will mainly take place not within the IEEE but the WiMAX Forum, which will work on improvements such as high speed mobility through soft hand-offs, and network management. But this group is highly susceptible to politics and delays, and if these are allowed to lengthen the timescales for early 802.16e products to any significant degree, it will find itself usurped by HSDPA and CDMA EV-DO, as both work on their own key weaknesses, such as indoor penetration, and start to eye the converged fixed-mobile market where WiMAX had hoped to be the unique contender.

_______________________________

This article originally ran in Wireless Watch, a publication of Rethink Research. Reproduced with permission. For information on the weekly Wireless Watch Newsletter and other Rethink Research products and services, click here.