It's something of a cliché, but physicist Niels Bohr summed it up best back in the 1930s: prediction is tough, especially when it's about the future.
But that's probably not what IT professionals looking to invest in high-speed wireless services want to hear. The wireless market is already a patchwork of access technologies, platforms and protocols, from Wi-Fi to mesh networks and 3G services. In the next few years, the arrival of services such as ultrawideband (UWB) will make the market even more complex. How are IT managers supposed to know which technologies will survive, which will fall by the wayside and which might never deliver?
The answer is that all of these technologies are likely to survive in some form, says Ken Greene, European technical director at wireless service provider iPass. In the future, mobile workers will access networks that seamlessly incorporate a range of access technologies, Greene believes. "Ask me how I see business mobility in five years, and I'd sum it up as unconscious connectivity," he says.
Today, iPass uses technology that allows wireless consumers to access hundreds of Wi-Fi hot spots in hotels and other public sites using a single password and account -- even if the hot spots are run by different providers and on different technology platforms. In the future, Greene says, this type of seamless roaming will be extended not just across Wi-Fi services, but also networks using anything from WiMax in the city and UWB in the home.
Low upfront costs
By 2008, a worker could open their laptop and automatically be connected to the fastest, cheapest wireless network. A connection would be established without the need to install any software or perform complex configuration tasks. "The user will, or should be, a lot less aware of the technologies he's using. He just cares about the service," says Greene.
That's good news for workers accessing public wireless networks, but what about enterprises looking to wireless-enable their own employees? To some extent, analysts say that it doesn't matter too much which wireless access technology is deployed -- the cost of today's wireless systems is so low that if a technology becomes obsolete, you're unlikely to lose out, says Caroline Sceats, an analyst with Forrester Research. "The risks involved in Wi-Fi and Bluetooth, for example, are low because the upfront costs are relatively low," she says.
Building a Wi-Fi hot spot, for example, can cost as little as £500 (although it's not unusual for a business to spend around £10,000 on a company-wide, secure wireless LAN). "Given the cost involved, does it really matter if the standard you buy today isn't around in three years?" says Ken Dulaney, of analyst firm Gartner Group. "The gains in productivity over 12 months more than outweigh the investment you'll make in the technology."
Businesses can also take comfort from the fact that transforming today's wireless technologies into a seamless, converged wireless communications infrastructure won't happen overnight. "Changing telecoms is like changing your bank account," says Paul Meakin, portfolio business development manager at networking supplier Damovo. "There's an in-built conservatism about the hassle, and the hype is often undermined by the reality."
But getting the best from any wireless access technology depends on choosing the appropriate product and deploying it in the right place. Although Wi-Fi might seem the most obvious choice because of its prevalence, there are alternatives. Here, ZDNet UK provides a rundown of some of the current -- and future -- frontrunners in wireless access:
FSO: Free Space Optics (FSO) has been around for almost 20 years and has been used for many years to send information from the pit to car technicians during Formula One races. Where radio sends information as radio waves across the megahertz (MHz) spectrum, FSO sends information as light across the terahertz (THz) spectrum.
The main advantage of FSO is that it costs a tenth of equivalent wired access --- around $20,000 for a typical large enterprise customer. It's also extremely fast, offering speeds of up to 1.2Gbps with promises of up to 2.4Gbps to come. Because the THz spectrum is unlicensed, carriers can get systems up and running extremely quickly. For this reason, FSO has been most popular in cases where buildings are close together but can't be physically linked -- for example, college campuses where buildings are separated by a river or main road.
However, FSO technology has some limitations because it is a line-of-sight technology. The transmitter and the carrier's receiver need to have an uninterrupted view of one another. If there is fog then fine water particles can disrupt FSO laser beams and shorten their reach. If there is high wind, buildings can sway and knock lasers out of alignment. Even birds flying close to laser transmitters can disrupt the signal.
MAN: Metropolitan Area Networks are essentially wireless wide area networks that are based on the IEEE's 802.2 standard. MANs are being rolled out in the UK by the likes of Cisco and Intel, which believes that WiMax and metropolitan networks will complement Wi-Fi to provide complete wireless access for urban workers connecting to LANs.
Metro networks offer blanket access to the Net throughout an area the size of a large city at speeds of 1.5 megabits per second -- compared to around 2Mbps for a typical Wi-Fi hot spot -- which functions only within a small area. MANs work by sending a signal from a cellular transmission site, which consumers access through a special chip in mobile devices.
Mesh radio: In conventional wireless networks, there are two basic models. First, a single base station transmits a signal to a single receiver (point to point) or a single base station transmits a signal to many receivers (point to multipoint). In a mesh network, a base station transmits to another base station, and the signal 'hops' to the next station in a relay (point to point to point). It's possible to cover an entire city quickly using mesh technology. It is also cheaper than building fibre, and offers speeds of up to 10Mbps -- not up to fibre, but significantly faster than DSL or cable solutions. However, there are some issues with mesh -- it's difficult to predict the strength of signal on mesh networks and it can be unreliable.
RFID: Radio frequency identification chips --- sometimes called smart tags -- work by emitting a radio signal that can be read by an electronic base station from up to 300 feet away. For example, Tesco attaches RFID tags to selected high-value items in its warehouses. Each tag emits a radio signal which is picked up by the base station and transmitted over a dedicated Radio Frequency (RF) network to a back-end database. This data can then be used to keep a close eye on stock levels. When they are combined with MEMS technology (micro-electronic mechanical systems), tags can also track their physical environment, including temperature -- making them invaluable where stock is perishable or where an audit trail is necessary.
Satellite broadband: Satellite broadband offers DSL-level service for around £60 per month, and is offered in the UK by BT, Tiscali and Aramiska. The advantage of satellite broadband is that you don't need to dig up roads or lay expensive cables. Moreover, although it's nominally a line-of-sight technology, so long as you can see the sky, chances are that you can receive a satellite signal. The technical issues around satellite broadband -- latency and a lack of standards between vendors -- are being addressed and two-way, high-speed services are available with speeds up to 2Mbps.
UWB: In homes and small offices, analysts predict a big future for ultrawideband (UWB), which allows data to be transmitted at a rate of one gigabit per second -- nearly 10 times faster than today's best Wi-Fi connections. Although UWB has a much shorter range -- 30 to 60 feet versus 100 to 200 feet for Wi-Fi -- that could be enough for many uses. For instance, UWB could replace USB (universal serial bus) cords that now connect computers with peripherals such as a keyboard and printer. Businesses could also use UWB to provide fast access to bandwidth hungry applications -- for example, looking up x-rays in hospitals. Overall, the market is expected to grow from zero to nearly six million UWB nodes embedded in various devices by 2007, according to tech consultancy In-Stat/MDR.
WiMax: Fixed wireless access, 802.16, is the technology behind new WiMax services, which will be launched later this year by the likes of Intel. Although fixed wireless access has been around for a few years, WiMax provides a standard for this type of wireless broadband access, and as a result products should be significantly cheaper and more reliable.
How does WiMax work? Imagine a beam of bandwidth that supports data-transmission rates of up to 70 megabits per second within a 30-mile radius of a WiMax antenna. Initially, WiMax will be used by broadband providers -- such as phone companies that offer digital subscriber line (DSL) service -- for last-mile delivery of broadband to consumers' homes. WiMax could potentially extend the reach of broadband to consumers who live too far from the exchange or in remote areas that are not suitable for DSL service.
In the future, WiMax vendors hope that the technology will also be able to communicate with a special chip in wireless devices, along the same lines as today's Centrino chip. The technology would be aimed at mobile workers who want to check email on the road. The combined market for 802.20 (MAN, see above) and WiMax hardware should reach about $1.5bn by 2008, according to ABI Research, a US-based research firm.
WozNet: WozNet is a proprietary wireless access technology developed by Wheels of Zeus, a company started by Steve Wozniak, co-founder of Apple. Wheels of Zeus says that it will release a new product in 2004, called wOzNet. Expected to be resold by Motorola, this specialised hardware will combine Wi-Fi-type access and GPS (global positioning system) technology to offer combined wireless and location-based services, so people can keep track of where their children, elderly relatives, or pets may be.