Controversy is in the air. Quite literally. It is being put about that third generation (3G) mobile technology is hugely expensive and will take several more years to get into the market. In the meantime, 2.5G technologies like GPRS (General Packet Radio Service) and HSCSD (High Speed Circuit Switched Data) together with wireless LANs are here now. As such they represent potential challengers for 3G at much lower cost – some $3bn as opposed to $250bn across Europe. This is the conclusion from a new report by investment bankers Merrill Lynch this month.

There are others too. Finnish ISP Jippi’s subsidiary Wireless Network Services, for one, has already built a network of public access wireless LANs in Finland. These offer wireless broadband Internet access for a flat monthly rate of about half the equivalent DSL price. Now the company is reported to be in discussions with mobile operators to offer roaming services later this year between GPRS networks and WLANs in Finland. At a fraction of the cost of a 3G platform too.

Then there’s Bluetooth. After what seemed like a rather slow start, this looks an increasingly promising wireless technology, is cheap and is rapidly coming onstream now. Originally this was envisaged to replace peripheral device cabling, like PC connections to printers and mobile handsets, with low cost wireless links. However, over 2000 companies have now signed up to use the technology, over a huge variety of applications including cordless phones (to replace DECT) and public "hotspots" for wireless PC access. Maybe Bluetooth can grab some of the revenue that 3G services were supposed to get as well.

Who needs 3G then? By the time it arrives, will all the big revenue opportunities have been creamed off by other technologies?

Rating the data rates

GPRS and HSCSD are essentially alternative techniques for enhancing GSM-based bandwidth by combining voice channels. HSCSD, by combining up to eight channels, can deliver 64kbit/s transmission speeds. GPRS, by combining channels and using new coding techniques reminiscent of 56kbit/s modems on fixed networks, can boost transmission rates to 115kbit/s – and possibly higher.

These data rates are not per user, though, they’re per cell so all users in a cell must share the full rate. Further, by combining channels, they reduce the voice capacity available in that cell. Not good where voice channels are already in short supply, such as in busy city centres and airports.

Worse, though, is the likely return on using the bandwidth for higher speed data rates. Can an HSCSD channel comprising, say, four voice channels and delivering 32 kbit/s be priced at four times the equivalent voice call? Not for long calls, one suspects, and certainly not for long international ones. GPRS uses different techniques for combining channels making it more suitable for bursty traffic but essentially the same economics apply – lower data rates are much more cost effective to provide than higher rates.

The hotspot supplement

Hotspots are places where high speed wireless data rates are most likely to be required. This particularly includes airport executive lounges but also hotel reception areas and conference facilities and, perhaps, railway stations and even shopping malls. These are the most likely locations where mobile PC users will want to sit down and access or download large amounts of data.

At least four alternative technologies are shaping up to compete in this area, with Bluetooth and IEEE 802.11 wireless LAN the likely chief contenders, with each offering significantly different data speeds These provide the basis for wireless hook-up points, themselves connected to local area networks in the building. Mobile PCs can then be connected wirelessly to these in-building LANs, which in turn provide fixed line access to the Internet and whatever corporate intranet the user needs. Such services could be managed directly by the building owners themselves or on their behalf by an external service provider, inevitably competing directly with 3G.

There is a snag though. Both technologies use the 2.4 GHz band which is license-exempt in all European countries except the UK, where there are strict distinctions between public and private network licenses. As it currently stands, this means that operators in the UK are not allowed to charge for services delivered over this band.

From a business perspective too there are difficulties in establishing the economics of service provision from a variety of different types of location. In the US, three airlines are currently vying to be the first to offer travelling executives the opportunity to hook up wirelessly in their executive lounges. This is not part of a revenue-generating business model, though. It’s simply a new competitive offering subsidised by business class ticket sales. There is a real prospect that the hotspot business case could become too complex to co-ordinate the different vested interests in each type of location – rather like the Telepoint cordless telephony services of the 1980s, which were never able to demonstrate a profitable return.

Filling the 3G gaps

There is no doubt that these technologies offer some exciting product and service opportunities. They may also take over some of the 3G revenue opportunities.

However the likelihood is that they will in fact help significantly to handle areas where 3G operation was a bit questionable. An integrated 3G/Bluetooth handset, for example, will quickly be able to incorporate many applications that 3G alone would find difficult to achieve economically. Bluetooth may then well make the 3G handset more of a must-have than originally envisaged. Wireless hotspots, too, are more likely to be complimentary than compete directly with 3G. What, after all, will mobile PC users connect through in all those places where hotspots are uneconomic? Initially 2.5G, but with the disparity of speeds available at hotspots and through 2.5G networks it won’t be long before 3G looks more attractive.

What is much more likely, then, is that these technologies will help to create demand for 3G when it finally arrives.

© e-principles 2001

Robin Duke-Woolley

Any comments on this article? Please send them to : Editor@e-principles.com