BLUETOOTH - The tooth is out there

Expect the unexpected. The most unlikely partners, from mobile phone vendors to PC makers, have stated their combined support for the latest data interface standard.

It's not every day a cross-section of industry players unites tohone vendors to PC makers, have stated their combined support for the latest data interface standard. promote the same technology. But at the beginning of the summer, Intel, IBM and Toshiba joined forces with key mobile handset manufacturers, Nokia and Ericsson, to create a data interface standard codenamed Bluetooth.

Bluetooth is painted by its developers as a huge boon for the portable PC user, but some industry observers argue it is little more than a ploy by Intel simply to sell lots of Pentium processors into the portable computing area, where Risc-based 'smart' handsets are beginning to mount a challenge.

Although the initial push behind Bluetooth has come from these five corporates, together with the palm computing division of 3Com, the initiative has attracted much wider industry support. The list of backers includes Compaq, Dell, Lucent and Motorola, as well as companies that would be expected to oppose the move, such as VLSI and Hewlett Packard.

Significantly, if Bluetooth is widely adopted, it will sweep away the myriad of proprietary data interfaces currently restricting the growth of data over cellular systems. This will certainly make wireless systems much easier for the IT dealer community to resell.

Bluetooth's objective is to provide a simple but intelligent means of exchanging data between two portable communications devices, particularly mobile phones and portable PCs. Its key design features are its ability to work in all global markets, its negligible power consumption and a small footprint.

According to a representative at Toshiba UK, there is enough real estate on the vendor's portable PC motherboards to accommodate Bluetooth and a rival technology such as an IrDA port - IrDA is the infra-red standard for data transfer between devices formulated by the Infra-red Data Association - with ease.

The alliance also aims to make the technology cheap to implement and is positioning it as an open standard. US IT consultant Andrew Seybold predicts Bluetooth chips will initially cost between $15 and $20, but says by 2001 they should fall below $10 to as little as $5 per chip.

If everything goes according to plan, Bluetooth will effectively displace IrDA. The IrDA ports have already made a reasonable showing among mainstream notebook PCs and PDAs in particular. They have even started to appear in some printers, such as those from Hewlett Packard, and the odd mobile handset such as Nokia.

The great advantage Bluetooth has over infra-red is that it is radio-based and so works without line of sight, eliminating the restrictive need to 'point and shoot'. This means that, theoretically, a mobile handset in someone's jacket pocket should have no problem interfacing with a portable PC stored in a briefcase on the back seat of their car.

Added to this, the Bluetooth technology enables two devices to synchronise themselves automatically when they are within range of each other. This means they will compare their databases and replicate any missing information according to a set of predefined rules. The feature is a result of the involvement of 3Com/Palm and Puma Technologies, which produces the IntelliSync range of data interchange applications.

The decision by the Bluetooth camp to handle wireless physical connection using radio technology rather than infra-red has caused some confusion.

But suggestions that Bluetooth presents a direct threat to existing digital cordless handset technologies such as DECT (digital enhanced cordless telephony) have been dismissed by Nokia and Ericsson. The manufacturers describe Bluetooth simply as a 'virtual cable' between two intelligent devices.

Others in the IT industry are playing down the handset-to-PC aspect too, arguing Bluetooth is a form of wireless Ethernet. Chris Russell, European director of marketing at chip manufacturer Cirrus Logic, says he welcomes the Bluetooth initiative. 'The addition of cost-effective RF (radio frequency) links to PCs and other consumer products will enhance the usefulness of technology in the home,' he adds. Since market research shows that the number of multiple PC homes will grow faster than single PC homes, the Soho market for networking products is set to show explosive growth.'

The catch, according to Russell, is that home PC users have proved reluctant to wire up their houses with Ethernet. 'To date, radio technology for wireless networks has been far too expensive for the broad mass of users to consider, but the emergence of a high-volume market will drive the cost down quickly,' he predicts.

The Bluetooth consortium is just one group that thinks it has found an answer to home networking. The electricity company Norweb, for example, is approaching the problem from a different tack - its solution involves squirting data down a mains cable.

In addition to this, Bluetooth is not the sole existing proposal for radio-based home Lans. Rival groups include the Home RF Working Group and the IEEE 802.11 Wireless Lan group. 'How Bluetooth's efforts may conflict with or complement those of other consortia remains to be seen,' says Russell.

Apart from the home market, Bluetooth's potential impact is significant and it could steal market share from existing wireless Lan adaptor suppliers that support business users. But Paul Sherry, European sales and marketing director at Netwave UK, which was recently acquired by Bay Networks, claims it does not pose a threat because its distance is restrictive - between 10 and 30ft, which isn't sufficient for office and factory use.

Sherry says: 'It's good news for us because it increases the trend for wireless to move from its current niche market status to the mainstream sector. We're already starting to see interest from the corporate-style accounts that want our product for the typical mobile professional type of user.'

Interest in wireless Ethernet was building pre-Bluetooth. Sherry adds: 'The market has begun to react to the introduction of 802.11, particularly now we have product that conforms to that spec. The most crucial aspect of 802.11 is that it enables corporates to buy from multiple suppliers - it makes wireless Lan products vendor-independent. The same should happen with Bluetooth, which will be used for a host of applications.'

There is potential for interference between Bluetooth devices and wireless LAN adaptors, particularly since they both use the same frequencies. However, Sherry is confident the issue is not a problem. 'There are already multiple wireless Lans using the same air space on the market,' he says.

Devices use SSDs - unique identification numbers - to ensure they communicate with the correct base station. Sherry argues that since Bluetooth will be based around the 802.11 standard, the situation will be resolved as it is at present.

Added weight for Bluetooth comes from Intel, which is waving the flag for the technology. Stephen Nachtsheim, Intel vice president for the mobile and handheld products group, states: 'Intel is pursuing a number of complementary wireless technology programmes and systems to address the unique requirements of different environments. To this end, it is engaging with and uniting leaders from the mobile computing, telecoms and peripherals industries to form the Bluetooth consortium.'

Bluetooth is just one of many attempts by Intel to kick-start the data over wireless market on a global scale. At $5 per chip, Intel's interest does not lie in breaking into the 'data interface' chip market, but in trying to defend its core OS processor business.

According to John Hughes, vice president of EMEA at Lucent Technologies: 'Lucent will be the first to supply the entire Bluetooth Radio, using advanced integrated circuit processes developed at Bell Laboratories.

In addition, Lucent's worldwide leadership in ASIC, DSP and embedded memory will enable best-in-class systems to be offered across the full range of Bluetooth applications.'

Andrew Till, technology strategist at Psion Dacom, claims Intel's real worry is that computing and telephony will be merged into one, similar to what Nokia has done with its 9000 Communicator range.

Rather than using Pentiums, this generation of intelligent mobile phones is using low-power Risc chips instead. So if a low-cost, portable computer to mobile handset technology is widely adopted, users will be less inclined to ditch their portable PCs and PDAs, many of which have Intel chips.

The implications of Bluetooth haven't been lost on manufacturers specialising in PC-to-handset systems either. PC card comms specialist TDK Grey Cell Europe has added its signature to the 18-strong consortium. TDK managing director, David Heppe, says: 'We want to be seen to be leaders in GSM data and we see the Bluetooth project as one of the next important steps, maybe second only to the speed enhancements coming later this year. We wanted to be there early, to participate in the Bluetooth committees as the standards are finalised to ensure users get the right technology as quickly as possible.' Industry observers expect Bluetooth-based product to be commercially available in the third quarter of next year.

Handset vendors hope Bluetooth will plug a gap that IrDA never managed to. Dr Yrjo Neuvo, senior vice president at Nokia Mobile Phones, concedes: 'Nokia has been developing short-range wireless connectivity for several years. The introduction of Bluetooth is a momentous landmark uniting the telecoms and computing industries through support of a common interface.' The hope is that Bluetooth will be implemented uniformly - something that has not happened with IrDA.

Most of the players that stand to be affected by Bluetooth are bullish about it. Russell says: 'Whatever the outcome, Cirrus is well positioned to take advantage, being one of few companies with core skills in advanced mixed signal processing, ARM-based controllers and algorithm development.'

Some observers believe Bluetooth's greatest rival will be DECT. Philips, for example, has already demonstrated how DECT can be used for PC data transfer with its Virtual Cable product. With the increasing popularity of DECT for home usage, the cost of the associated chips will be driven down as well. Samsung recently launched a DECT handset/base station package retailing for less than #100.

But Bluetooth has pulled in industry-wide support. Compaq and Dell on the PC side and Motorola and Qualcomm in handsets have endorsed it, even though their rivals are Bluetooth's original backers. It is most likely to go the way of PCMCIA, which was also touted as a solution for a variety of devices but is now used mainly for portable PC applications.

Bluetooth's chief merit will be that it will save dealers from having to carry a huge stock of different PC-to-mobile phone cables - even though those cables carried healthy margins.

HOW IT TRANSMITS

Bluetooth will operate in the unlicensed 2.45GHz region, allowing international travellers to use their equipment worldwide. It bears more than a passing resemblance to wireless Lan technology since it is based on the IEEE's 802.11 standard for wireless Ethernet. However, Bluetooth is restricted to a range of just 10m (about 40ft).

'It can transmit and receive through walls and doors and maintain an uninterrupted connection in motion,' claims Anders Aarsoe, mobile product sales manager at IBM. 'Bluetooth chips can transmit data at 1Mbps. The alliance plans to increase it to 2Mbps for the next generation, although no date has been set.'

The 1Mbps transmission speed is normally subdivided between all the devices within a particular area (called a piconet), so users are actually only guaranteed 720Kbps. Worse still, that's 600Kbps in one direction and 120Kbps in the other. On the plus side, Bluetooth chips only consume about one milliWatt.

Compared with Bluetooth, the typical 802.11 Lan adaptor has a range of 150ft and a data rate of between 1.6Mbps and 2Mbps. Furthermore, it is possible for such devices to work up to several miles away in open country.

The greatest difference is that an entry-level wireless Lan adaptor costs $149, compared with the Bluetooth pricepoint, which is expected eventually to hit $5.

'The nearest comparative technology is IrDA but that doesn't come close to Bluetooth's capabilities,' says Aarsoe. In fact, IrDA has a radius of between 2m and 3m and requires the user to point the transmitter directly at the receiving device.