IF YOU live out in the sticks, too far from your telephone exchange to have access to a broadband internet connection, you can soon expect to be hearing a lot about a technology called WiMAX.
By fixing a small receiver dish to your roof, and plugging it into a modem, the technology can provide broadband internet access from an antenna up to 10 kilometres away. Paid-for WiMAX services are already on offer to tens of thousands of houses in parts of Germany, France and Spain. Delivering broadband over the airwaves, rather than down telephone wires, has obvious appeal in rural areas and in developing countries that do not have wired infrastructure. Now a group of companies led by chip-maker Intel is planning to extend the concept with a version of WiMAX that can communicate with mobile devices such as laptops and smart phones.
But not everyone in the industry welcomes this development. Some say the claims being made for WiMAX鈥檚 performance and availability are exaggerated, and that third-generation (3G) cellphone technology can do the job just as well. They fear that WiMAX could become dominant purely by dint of its developers鈥� marketing might, jeopardising the even faster fourth-generation technology that鈥檚 in development.
Advertisement
At issue is just what WiMAX is best at. Championed by the WiMAX Forum, a trade association that includes Intel, Microsoft and many other big names, it is a way of sending data over the air at claimed rates of up to 75 megabits per second, over distances up to 10 kilometres. To do this it uses microwave signals in frequency bands between 2 and 11 gigahertz. At the higher end of this range the microwaves behave like light, and the receiver and base station antenna have to have clear line of sight to each other. Below 6 gigahertz, they are more like a cellphone signal, so line-of-sight contact is not necessary. Which parts of the frequency spectrum WiMAX will use for mobile services has not yet been decided.
WiMAX comes in two main flavours. The first is the fixed broadband system now running in Europe, code-named 802.16d, which was ratified as a standard last year by the Institute of Electrical and Electronics Engineers. The second is 802.16e, the mobile version still in development, which should be ratified by March 2006, with products promised for late 2006.
The 802.16d standard replaces a mass of different proprietary systems. The hope is that standardisation will fuel competition, which in turn will drive down equipment prices and make internet access more affordable to developing countries. According to Gilles Karolkowski, who runs Intel鈥檚 WiMAX operation in Paris, dish antennas will soon not be needed for the 802.16d system: from late next year receivers will be able to operate with a new type of smart antenna to be built into the modem.
It is the mobile version, 802.16e, which is ruffling feathers. The idea is that a new generation of WiMAX-capable laptops and smart phones will be able to maintain wireless broadband access even when they are many kilometres from a base station, unlike today鈥檚 Wi-Fi, which has a range of the order of 100 metres. Gone will be the days of having to find a coffee shop or library to get connected. As long as your mobile gadget is in range of a WiMAX transmitter, you will be able to get online at high speeds.
鈥淔irms are not keen to correct the perception that 70 megabits a second is available鈥�
That, at least, is what WiMAX advocates claim. But 802.16e is now the target of growing criticism because the WiMAX Forum鈥檚 promises are starting to seem a little optimistic. 鈥淲iMAX reach and bandwidth claims are totally out of whack,鈥� says Charles Golvin of technology analysts Forrester Research in Boston. Calculations by the French telecoms company Alcatel show that the promise of 70 megabits per second is flawed. Once the signal has been shared by a typical number of users in a WiMAX cell, each gets somewhere between 512 kbit/s and 1.54 Mbit/s, Golvin says. The calculations are borne out by the bandwidth delivered by the fixed WiMAX systems up and running in France and Germany: 1 Mbit/s is available from Altitude Telecom in France and 1.5 Mbit/s from Deutsche Breitband Dienste in Germany. Forrester researchers accuse WiMAX Forum members of being 鈥渦ninterested in correcting the widespread perception that 70 Mbit/s is available to the end user鈥�.
At rates of around 1 Mbit/s, WiMAX offers little that a user with a 3G smart phone, or a laptop with a 3G modem card, won鈥檛 soon be able to achieve when base stations are equipped with software called High-Speed Downlink Packet Access (HSDPA). Today, 3G users typically get 300 kbit/s to play with. By squeezing more data bits into each radio pulse HSDPA, sometimes called 3.5G, will boost that to 1 Mbit/s or more. Users will need to upgrade their handsets and modems to receive the signal.
Regulatory mess
The O2 network is about to roll out HSDPA on its 3G network on the Isle of Man, and new handsets and HSDPA cards will be available to make use of it. Mike Short of O2 says he expects laptop computers with built-in HSDPA 3G receivers to be released in early 2007 by firms such as Fujitsu-Siemens.
That will be around the time that devices with built-in Wi-Fi and WiMAX mobile capability will also be appearing. But analysts say that problems freeing up the necessary microwave frequencies will seriously slow development of WiMAX mobile hardware. For people to be able to roam across different countries with WiMAX mobiles, the frequencies must be on the same channels in each, and it will be several years before this can happen. 鈥淓urope is a regulatory mess when it comes to frequencies,鈥� says Golvin. Ian Keene of the market research group Gartner, based in Stamford, Connecticut, is projecting that the frequency problem won鈥檛 be solved until 2009 or 2010, by which time 3G will be headed for its 4G incarnation. NTT DoCoMo in Japan has already demonstrated 4G technology at speeds of 100 Mbit/s.
Another problem with WiMAX is uncertainty over whether it will ever truly be mobile in the way that cellphones are. A spokeswoman for Nokia, which is working with Intel on WiMAX, says it may offer 鈥渓imited mobility鈥�, allowing users to take a portable gadget around with them and then use it while stationary. Whether the system can be made truly mobile, so that it can be used on a moving train, for example, will ultimately depend on the level of sophistication of the technology used to hand over devices from one cell to the next, and the number of WiMAX cells provided within a given area.
Handover will be especially challenging for WiMAX because, unlike 3G, it will probably have to switch between a broad range of frequencies. 鈥淲e have our doubts that WiMAX can truly compete with 3G,鈥� Keene says. 鈥淭here are just too many ifs and buts about the technology鈥檚 development right now.鈥� Intel, however, insists that WiMAX has its place. 鈥淲e think 3G is the best option for voice, while WiMAX is the best for data delivery. In addition, 3G is not useful for large numbers of users, like WiMAX,鈥� Karolkowski says.
Whatever the problems of WiMAX, it may not be technical merit that determines its commercial success. Intel popularised Wi-Fi by adding a radio access chip to its Pentium processor and spending in excess of $200 million worldwide advertising the chipset. Keene and other commentators expect a similar marketing blitz for WiMAX. As Mark Twain said: 鈥淢any a small thing has been made large by the right kind of advertising.鈥�
