The good, the bad and the reality of Bluetooth by Laura Rohde, July 10, 2000, 11:37 Harald Bluetooth Thedanish (Personae) Born: 925 * Baptised: ~Okt 940 King of Danmark Died: >986 With the backing of some of the biggest names in computers and consumer electronics, the wireless voice and data technology known as Bluetooth has attracted tremendous media and consumer attention over the last 18 months. But as the launch of the first Bluetooth products looms in the fourth quarter, even its strongest proponents are worried that hype and competing claims about what it can do could cause user confusion or at worst even sink the technology before it sets sail. The media and consumers have flocked to demonstrations of Bluetooth technology at trade shows around the world, dozens if not hundreds of companies have fixed launch dates for products by now, and the main backers of the technology say that more than 2,000 vendors have at least expressed interest in incorporating the technology into products ranging from mobile phones to laptop computers to personal digital assistants (PDAs). Bluetooth has the potential to let all sorts of devices link up, and its proponents paint tantalizing pictures of how it could work. "You could have a pair of mufflers -- the headphones over your ears -- and be mowing the lawn outside, listening to your Walkman when your phone rings inside and automatically stops the music to tell you that there is a call which you can then take," suggested Christina Bjorknader, marketing and communications manager for L.M. Ericsson Telephone, a major Bluetooth backer. Her scenario continues: Depending on whether your PDA has voice capabilities, you could even check your calendar while on the phone call and add an appointment or make a change. "Once you're done with the call, you can tell the headset to hang up the phone, which will simply restart the music from where you left off and you then finish mowing the lawn, all without taking your hands off the mower," she said. But some Bluetooth proponents warn about overselling the technology. "People are beginning to get a little wild about what you can do with Bluetooth," said Nick Hunn, research and development director for TDK Systems Europe, at the recent Networks Telecom 2000 conference in Birmingham, U.K. Hunn, who has been involved with the development of the Bluetooth standard from the technology's inception, has also worked on incorporating Bluetooth into existing TDK products such as PC Cards, USB Adapters and network routers. "In these early days Bluetooth won't be cheap as companies try to recoup development costs. But it's very important that we don't raise expectations to begin with. On day one, Bluetooth is a cable replacement system: don't expect anymore at first," Hunn said The idea behind Bluetooth is to avoid the inconvenience of cables by enabling devices such as mobile phones, PCs, printers and handheld computers to communicate with one another over short distances using low-power radio signals to transmit data. Bluetooth operates in the 2.4GHz frequency band and the original Bluetooth specification calls for output power of less than 10 milliwatts. At events like Networks Telecom, vendors have stoked the hype for the technology. "My personal belief is that Bluetooth is the most exciting technology to hit this industry since GSM (Global System for Mobile Communications), especially since it is so easy to use," said David Lock, commercial strategy manager for U.K. mobile phone operator Orange, at the show. Bluetooth technology was originally designed in 1994 by two L.M. Ericsson Telephone employees, the Swedish born Sven Mattisson and his Dutch colleague, Jaap Haartsen. Though Ericsson's Bjorknader concedes that both men have expressed concerns that vendors may try and stretch the radio technology past its limits, she said Ericsson is "not so worried about the hype." "Ericsson hasn't seen anything yet that Bluetooth couldn't handle, which is the beauty of the technology," Bjorknader added. However, there has been some controversy over what exactly Bluetooth is capable of. For example, whether it would make a good wireless LAN technology is the subject of debate. Bluetooth does have the ability to provide a link on private networks -- dubbed "Piconets" -- with one master that takes up to seven slave devices. Bluetooth is billed as working at speeds of up to 721K bps (bits per second), but TDK's Hunn believes the reality will more likely be between 200K and 400K bps. Given the speed limitations, he said, most companies wanting to install new wireless networks will probably want to wait 18 months for HyperLAN2, "which is better than Bluetooth for video conference streaming anyway." The distance over which Bluetooth can work on a practical basis is also being debated. The technology allows "frequency hopping over a spread spectrum" and supports both voice and data. "Bluetooth divides the spectrum into 79 different frequencies and jumps over any interference it encounters," Hunn said. The technology is effective as a mobile router; it can route transmissions to the PSTN (public switched telephone network) and offers low-speed wireless LAN connectivity. And though Bluetooth was designed to work at short ranges of about 10 meters, it can work up to ranges of 100 meters. "I personally think this is a bad option because it makes things too crowded, rather like rush hour on (the U.K. highway) M25," TDK's Hunn said. Also, he pointed out, Bluetooth uses unlicensed spectrum and will get interference from radio frequencies used by 802.11 and 802.11b wireless networking specs/protocols, from cordless telephones using the DECT (digital enhanced cordless telecommunications) standard, as well as from microwa= ve ovens and some light bulbs using long-life technology. Ericsson's Bjorknader does not believe that a crowded spectrum will pose any problems for Bluetooth due to its ability to achieve 1,600 hops per second. "Going up to 100 meters will not create more interference for Bluetooth. We knew it was going to be a tricky problem but Bluetooth is going to be the one that wins (in frequency traffic wars) because it changes frequencies so quickly," Bjorknader said. But what is most important about Bluetooth is that the technology has been designed to be cheap, industry insiders agree. It currently costs about $25 for a company to incorporate Bluetooth into a product, but by the year 2004 or 2005, industry sources see that price dropping to about $5. Bluetooth analyst Nigel Deighton, from Gartner Group, sees the low price point being key for Bluetooth's acceptance on the mass market. "Assuming that hardware vendors follow through on their promise of $5 to $6 chips in a couple of years, it will be a popular consumer product. I expect Bluetooth to hit the mass market in the first half of 2002," Deighton said. The first five companies to officially back the Bluetooth technology with money, Ericsson, IBM, Intel, Nokia and Toshiba, created the Bluetooth Special Interest Group (SIG), and later the nine-member Bluetooth Promoter Group. The Bluetooth Promoter Group includes 3Com Corp, Lucent Technologies, Microsoft and Motorola, and focuses on improving the Bluetooth specification and conducting interoperability tests. According to the Bluetooth group the most popular scenarios will most likely be connecting the laptop PC to the mobile phone; universal headsets connecting such peripheral devices such as mobile phones and Walkmans to each other; sh= ort range data transfer; and synchronization of data residing in different devices (for example, PDAs and laptops). Earlier this month, mobile phone vendor Ericsson demonstrated a WAP (wireless access protocol Bluetooth mobile phone, the tri-band T36, as well as a Bluetooth PC Card, which allows a user to connect a laptop to a mobile phone using the Bluetooth wireless connection. The Swedish company said the products would hit the market by the fourth quarter. Analysts such at Deighton agree that initially Bluetooth will be taken up by corporations with heavy network users who are looking to take advantage of the convenient technology as well as lose their tangle of cables. "The heavy corporate users will use the PC cards and the mobile phones." The Bluetooth group also sees lucrative opportunities between Bluetooth and GPRS (General Packet Radio Services) high-speed network technology. GPRS enables mobile devices to remain permanently connected to the Internet and transmit data at five to 10 times the speed of current transfer rates of dial-up modems at 56K bits per second (bps). It also speeds up Bluetooth transmissions. While Ericsson's Bjorknader predicts that Bluetooth will be a global technology straight away, TDK's Hunn believes that Bluetooth will be most important in the European market because of its dominance in wireless and mobile technologies. "For every laptop in Europe, there are 20 mobile phones. Europe is going to drive the growth in Bluetooth because of the mobile. Bluetooth will very much be a European-centric market," Hunn said. Because of the limited mobile spectrum in Japan, Bluetooth in handheld devices is being delayed there until the launch of third-generation (3G) wireless networking technology, in 2003 or 2004. "Because the Japanese don't wan= t any further use of the current spectrum, what will drive Bluetooth in the Japan market are consumer and sub-consumer notebooks," Hunn predicted. The U.S. may need GPRS and 3G in order to join the Bluetooth growth curve, but again, that wouldn't happen until 2003 or 2004. "The U.S. is not a market driven by mobile technology and won't be for the next two or three years and because of that I think that (data) synchronization may be the first important use of Bluetooth there," Hunn said. For the time being however, as Bluetooth becomes a reality, practical issues related to compatibility of products using slightly different Bluetooth specifications are raising concern. The initial standard was developed before any products existed and as developers began putting Bluetooth into devices, they invariably found bugs. As a result, there has been one revision of the Bluetooth standard called 1.0b so far, and there is another standard revision called 1.1 due in September that incorporates some 76 major changes to the standard. But in the rush to get Bluetooth products out on the market place, companies are using the 1.0b standard and as a result, "those products won't talk to 1.1 standard -- using products released just two months later. I just hope it doesn't kill the market before it gets there," TDK's Hunn said. Gartner's Deighton said he wouldn't be surprised if the different Bluetooth standards caused compatibility problems between devices but didn't expect incompatible Bluetooth products to be such a problem. "I would expect there would be backward compatibly built into the new devices," Deighton said. Ericsson's Bjorknader believes that the main concern for Bluetooth will be the number of products that carry the chip and work with the technology. "We've got to make sure that we have a lot of Bluetooth products out on the market to insure interoperability. Those products are going to come in a few years but we do need more companies to announce that they are working on upcoming Bluetooth products," Bjorknader said. The bottom line, according to Hunn, is that Bluetooth should not be oversold. Because while it is not, for example, a true wireless LAN, its potential as a very good cable replacement technology is sufficient reason to buy it. "It makes things work easier, which is why there are so many reasons to buy Bluetooth," Hunn said. Ericsson, in Stockholm, can be reached via the Web at http://www.ericsson.com/. TDK Systems Europe, located in London, can be contacted at http://www.tdkgreycell.com/. Orange, in London, is at http://uk.orange.net. The Official Bluetooth Web site can be found at http://www.bluetooth.com/. Copyright (c) 1999 International Data Group. All Rights Reserved Bluetooth technology and implications - Page 1/3 Article Date: Dec. 14, 1999 By: Heidi Monson What is Bluetooth? Think of it as an eighteen-wheeler screaming down the pike, carrying the future of computing with it. The driver is friendly. You can stick out your thumb and hitch a ride, or be left in the dust. Bluetooth is a developing, world wide, open, short-range radio specification focused on communication between the Internet and Net devices, plus it defines communication protocols between devices and computers. To be Bluetooth certified, a device must pass interoperability testing by the Bluetooth Special Interest Group (SIG), thus assuring that products meeting the specification will be able to interact with all other Bluetooth-certified products and with the Internet. The five founding members of the Bluetooth SIG are Ericsson, IBM, Intel, Nokia and Toshiba. On December 1, 1999, the founding members announced that 3COM, Lucent, Microsoft, and Motorola have joined the founders to form the Promoter Group with the founding companies. To date, more than 1,200 companies have signed on. Picture the spaghetti of wires connecting peripherals to a PC, and the PC to the Net, and then imagine them gone. Nice image, isn't it? The technology of Bluetooth centers around a 9mm x 9mm microchip, which functions as a low cost and short range radio link. It provides security for both stationary and mobile devices. The basic function is to provide a standard wireless technology to replace the multitude of propriety cables currently linking computing devices. Better than the image of the spaghetti-free computer system is the ability of the radio technology to network when away from traditional networking structures, such as a business intranet. Even better is the ability to network on an ad hoc basis. For example, imagine being on a business trip with a laptop and a phone. Bluetooth technology allows interfacing the two. Then, picture meeting a client and transferring files without cabling or worrying about protocols. That is what Bluetooth will do. The name game: whose teeth are blue? After exhaustive research (and more importantly, a friend who just happened on the information), the etymology of the name Bluetooth can be reported here. It comes from a tenth century Scandinavian king, Harald Bluetooth, who managed to unite several unruly kingdoms. Thus, Bluetooth is a reference to the taming of a myriad of unruly competing standards by defining one world-wide specification. Overview of Bluetooth characteristics These are the features of the Bluetooth technology: It separates the frequency band into hops. This spread spectrum is used to hop from one channel to another, which adds a strong layer of security. Up to eight devices can be networked in a piconet. Signals can be transmitted through walls and briefcases, thus eliminating the need for line-of-sight. Devices do not need to be pointed at each other, as signals are omni-directional. Both synchronous and asynchronous applications are supported, making it easy to implement on a variety of devices and for a variety of services, such as voice and Internet. Governments worldwide regulate it, so it is possible to utilize the same standard wherever one travels. Bluetooth technology and implications - Page 2/3 Article Date: Dec. 14, 1999 By: Heidi Monson Bluetooth technology Bluetooth definitions Piconet: Devices connected in an ad hoc fashion, that is, not requiring predefinition and planning, as with a standard network. Two to eight devices can be networked into a piconet. It is a peer network, that is, once connected, each device has equal access to the others. However, one device is defined as master, and the others as slaves. Scatternet: Several piconets may form a larger scatternet, with each piconet maintaining independence. Master unit: The master in a piconet whose clock and hopping sequence synchronizes the other devices. Slave unit: Devices in a piconet that are not the master. MAC address: Three bit address that distinguishes each unit in a piconet. Parked units: Piconet devices that are synchronized but don't have MAC addresses. Sniff and hold mode: Power-saving mode of a piconet device. Network arrangements Bluetooth network arrangements (topology) can be either point-to-point or point-to-multipoint. Any unit in a piconet can establish a connection to another piconet to form a scatternet. See the figure to the right, which diagrams a scatternet in which piconet A, which consists of four units, is connected to piconet B, consisting of two units. Note that the master unit of A is not the link between the two piconets. Transmission types and rates The baseband (single channel per line) protocol combines circuit and packet switching. To assure that packets do not arrive out of order, slots (up to five) can be reserved for synchronous packets. As noted earlier, a different hop signal is used for each packet. Circuit switching can be either asynchronous or synchronous. Up to three synchronous (voice) data channels, or one synchronous and one asynchronous data channel, can be supported on one channel. Each synchronous channel can support a 64 Kb/s transfer rate, which is fully adequate for voice transmissions. An asynchronous channel can transmit as much as 721 Kb/s in one direction and 57.6 Kb/s in the opposite direction. It is also possible for an asynchronous connection to support 432.6 Kb/s in both directions if the link is symmetric. Radio frequency and spectrum hopping What if there's a lot of radio noise? Won't that interfere with Bluetooth connections? As a rule, the answer is no. It is designed to use fast acknowledgement and frequency hopping, which will make connections robust. It is packet-based, and will jump to a new frequency after each packet is received, which not only helps limit interference problems, but also adds to security. Data rates are one megabyte/second, including headers. Full duplex transmissions (both directions at once) are accomplished via time division multiplexing. The Bluetooth radio chip functions at 2.4 gigahertz, which is in the unlicensed ISM (Industrial Scientific Medical) band. It separates the 2.4 gigahertz frequency band into 79 hops one megahertz apart, starting with 2.402 and ending with 2.480 (though this bandwidth is narrower in Japan, France, and Spain). This spread spectrum is used to hop from one channel to another, pseudo-randomly, which adds a strong layer of security. Up to 1600 hops per second can be made. The standard frequency range is 10 centimeters to 10 meters, and can be extended to at least 100 meters by increasing transmission power. Connection protocol Bluetooth connections are established via the following techniques: 1.Standby: Devices not connected in a piconet are in standby mode. In this mode, they listen for messages every 1.28 seconds over 32 hop frequencies (fewer in Japan, Spain, and France). 2.Page/Inquiry: If a device wishes to make a connection with another device, it sends out a page message, if the address is known, or an inquiry followed by a page message, if it is unknown. The master unit sends out 16 identical page messages on 16 hop frequencies to the slave unit. If there is no response, the master retransmits on the other 16 hop frequencies. The inquiry method requires an extra response from the slave unit, since the MAC address is unknown to the master unit. 3.Active: Data transmission occurs. 4.Hold: When either the master or slave wishes, a hold mode can be established, during which no data is transmitted. The purpose of this is to conserve power. Otherwise, there is a constant data exchange. A typical reason for going into hold mode is the connection of several piconets. 5.Sniff: The sniff mode, applicable only to slave units, is for power conservation, though not at as reduced a level as hold. During this mode, the slave does not take an active role in the piconet, but listens at a reduced level. This is usually a programmable setting. 6.Park: Park mode is a more reduced level of activity than the hold mode. During it, the slave is synchronized to the piconet, thus not requiring full reactivation, but is not part of the traffic. In this state, they do not have MAC addresses, but only listen enough to keep their synchronization with the master and check for broadcast messages. Data transmission As noted earlier, data can be transmitted both synchronously and asynchronously. The Synchronous Connection Oriented (SCO) method is used primarily for voice, and Asynchronous Connectionless (ACL) is primarily for data. Within a piconet, each master-slave pair can use a different transmission mode, and modes can be changed at any time. Time Division Duplex (TDD) is used by both SCO and ACL, and both support 16 types of packets, four of which are control packets that are the same in each type. Because of the need for smoothness in data transmission, SCO packets are generally delivered via reserved intervals, that is, the packets are sent in groups without allowing other transmissions to interrupt. SCO packets can be transmitted without polling by the sending unit. ACL links support both symmetric and assymetric transmissions. Bandwidth is controlled by the master unit, which determines how much of the total each slave unit can use. Slaves cannot transmit data until they have been polled by the master, and the master can broadcast messages to the slave units via ACL link. Error correction and security Three error correction techniques have been defined: 1/3 rate forward error correction code (FEC), 2/3 rate forward error correction code FEC, and automatic repeat request (ARQ). The FEC methods are designed to reduce the number of retransmissions. However, the overhead significantly slows transmissions, so is generally not used in relatively error-free environments, with the exception of packet headers. The ARQ scheme requires that the header error and cyclic redundancy checks are okay. When they are, an acknowledge is sent. When they aren't, the data is resent. Security is provided in three ways: through pseudo-random frequency band hops, authentication, and encryption. Frequency band hops make it difficult for anyone to eavesdrop. Authentication allows a user to control connectivity to only devices specified. Encryption uses secret key lengths of 1, 40, and 64 bits. The quality of security is excellent for most applications. However, it is not the highest level available, and for those users who require it, the suggestion is to investigate separate network transfer protocols and security software. Control of link connections The basic part of the Bluetooth system consists of the radio chip and controller, as shown in the figure above. The Link Manager (LM) is software that controls link setup, authentication, link configuration, and other protocols. The hardware underlying the LM is the link controller (LC). These two perform the following tasks: Sending and receiving data. Paging and inquiries. Setting up connections. Authentication. Negotiating and setting up link types, i.e., SCO or ACL. Determining the frame type of each packet. Placing a device in sniff or hold mode. Bluetooth technology and implications - Page 3/3 Article Date: Dec. 14, 1999 By: Heidi Monson Is Bluetooth really the wave of the future? To be fair, it must be admitted that not all pundits agree on whether Bluetooth technology is truly the wave of the future. Mark Riseley, an analyst at Inteco, has said, "People are not crying out for Bluetooth devices. They will inevitably be asking, "Do I want to switch on the lights using my mobile phone?" It seems a bit like taking a hammer to crack a nut and may not be considered worth the money." Let's first look at the potential applications-a difficult task, considering the virtually infinite nature of them. So, we'll start with Riseley's point of view. He refers to applications like switching on lights with a mobile phone. Certainly, such applications are possible. However, that's a minor one. So many more have far greater practicallity. As this article mentioned earlier, the removal of the spaghetti of cords surrounding a PC is one that most non-technical (and technical) users can relate to. Of course, there are the esthetics of not having that junk around. Beyond that, though, there is more practical consideration for removing it: the danger associated with tripping or electrical shock from worn cables. When the typical user sees the advantages there-coupled with the likely low cost of Bluetooth devices-it seems likely that moving to products that provide the technology will seem like a good choice. However, it is not the average user who leads the way in acceptance of new technologies; it is the business person on the go, the traveler, who demands products to stay in touch with the home office and home, and to make giving presentations easier. These are the people who jumped on cell phones and PDIs. The convenience of checking for e-mail while stuck in traffic is a strong motivator. The ease of linking with a printer or fax at the client's office, which Bluetooth can provide, must be irresistible. To be able to give a presentation with a laptop and not have to deal with cables and plugging in peripherals is a dream come true. Then, there are the conveniences of gadgets able to do double duty. "Where did I leave the TV remote?" can become a question of the past when the cell phone that just happens to be in your coat pocket will do the channel switching just as well. Picture yourself in the car on the way to a client's office when you realize you've forgotten to bring a file that's on your home PC. Wouldn't it be wonderful not to have to turn around, go back home, get out of the car, boot up the PC, dig out a blank floppy disk, copy the file, and then head back out? Imagine dialing the Internet from your cell phone and downloading the file while you drive along. What about the competition? If, as seems inevitable, some sort of wireless standard is the wave of the future, consideration must be given to a competing standard, the IrDA (Infrared Data Association) specification. This specification also defines a wireless protocol. Its characteristics include: Already proven universal standard for wireless connections, with an installed base of over 50 million. An existing wide range of platforms. Existing design for cable replacement. Backward compatibility with previous standards. Narrow 30 degree angle connections, which helps assure a lack of interference with other devices, but limits ease of application. Fast data transfers, currently up to four megabits per second, with 16 megabits per second in development. It is currently being used in a wide variety of applications, such as notebooks, desktops, PDIs, printers, phones, pagers, modems , cameras, LANs, medical and industrial equipment, and watches. In other words, both Bluetooth and IrDA are applied to the same types of devices. So, why would Bluetooth be successful when there's already a successful wireless standard? The answer is that the two standards will likely coexist, as they have complementary strengths and weaknesses. The narrow, point and shoot, type of application, such as crowded situations in which two people wish to exchange information while others are trying to do the same thing, is a strength of IrDA. In such situations, a Bluetooth device, which is omnidirectional, would be at a disadvantage, as it might have trouble finding an individual device among many. However, Bluetooth has distinct advantages in other situations. It can make connections through walls, and once a connection is established, it isn't necessary for any of the devices to remain stationary. The cost of Bluetooth implementation is higher than for IrDA. Presently, it runs about $2.00 for IrDA, and will likely run around $20.00 for the first Bluetooth devices, but drop to $5.00 after the initial implementations. This is not, of course, an indepth look at IrDA. It should suffice, though, to say that there are advantages to both technologies. More importantly, it is not necessary to choose one over the other, as they can coexist on the same device, with the appropriate one activated depending on the circumstances. Thus, to present one as superior to the other would be inaccurate. It is better to say that awareness of the differences is a more important focus, and that both are likely to exist in the coming age of wireless communications. Implications for the PC world There has been some murmuring to the effect that the coming wireless world will spell the end of desktop PCs. Such an approach seems rather narrow-something of an either/or proposition. A wireless world (with the exception of power cords, still a necessary evil) could prove to be a boon to PCs. There is, of course, the convenience of no spaghetti-wired desks, but more importantly, it would make the PC even more usable. Getting data to a PC is often one of the more difficult tasks. Dealing with storage media and their myriad of formats will become a thing of the past. Having to juggle peripherals that must share ports will no longer be a problem. Getting data from a laptop to a desktop will be simple. Adding a new peripheral will be a simple thing, instead of the nightmare it sometimes implies. This writer, at least, can hardly wait.