[2 articles re: Bluetooth/802.11b]
Mobilian Corporation Introduces TrueConnectivity, an Innovative Vision for Seamless Wireless Connectivity.
PORTLAND, Ore.--(BUSINESS WIRE)--Oct. 30, 2000
Start-Up Develops First Solution to Solve Emerging Bluetooth and Wi-Fi (802.11B) Coexistence Issues
Mobilian Corporation, a new wireless systems company that designs and develops integrated circuits and software, today announces their TrueConnectivity(TM) vision and first product, TrueRadio(TM).
The TrueRadio(TM) solution is the first highly integrated multi-standard wireless solution that supports simultaneous operation (Sim-OPTM) of Bluetooth(TM) and Wi-Fi(TM) (IEEE 802.11b) standards.
The TrueConnectivity(TM) Vision
"Mobilian's TrueConnectivity(TM) architecture will provide client devices such as laptops and PDAs with the ability to automatically detect all available wireless networks, select the ones needed, and connect to them seamlessly and reliably," said Manpreet Khaira, founder and CEO of Mobilian. "While our TrueRadioTM solution addresses the existing needs of customers to connect to WLAN and WPAN, future solutions from Mobilian will support a variety of wireless connectivity needs."
In a wireless world, a mobile user will need to connect to three types of networks, WPANs (Wireless Personal Area Networks), WLANs (Wireless Local Area Networks) and WWANs (Wireless Wide Area Networks), which are differentiated by range, data rates, power consumption and cost. WPANs, such as Bluetooth(TM) piconets, are designed to provide short-range connectivity for peripherals such as mice, keyboards, and headsets, and support low data rates and limited ranges in order to achieve low cost and minimal power drain. WLANs, such as 802.11b-enabled wireless Ethernet, are designed to provide higher speeds and longer ranges in office buildings and homes. WWANs, such as cellular or PCS networks, are designed to work over a large area but offer much lower data rates than WLANs. Most common usage models and applications require simultaneous connections to two or more of these networks, which creates enormous problems for existing wireless solutions. By enabling users to connect to two or more of these types of networks simultaneously, Mobilian's solution makes true wireless connectivity a reality.
TrueRadio(TM) Products: Delivering "Coexistence without Compromise"(TM)
"Bluetooth(TM) and 802.11 will blast into the market over the next two years, and will converge in notebook computers. Bluetooth(TM) will provide connectivity to all manner of pocket size devices, and 802.11 will provide the range and bandwidth needed for LAN applications. Many computing devices will have to deliver seamless support for both standards, within existing platform form factors and without interoperability problems," says Martin Reynolds, Vice President and Research Fellow, Gartner Group.
Based on the patented TrueConnectivity(TM) architecture, TrueRadio(TM) products enable coexistence of BluetoothTM and Wi-FiTM in a highly integrated two-chip solution; an analog front-end chip and a digital baseband chip. The solution allows simultaneous operation of the two protocols while eliminating interference and maintaining reliability and performance. For example, a laptop user with a TrueRadio(TM) based wireless network card would be able to seamlessly connect to Bluetooth(TM) enabled wireless peripherals while simultaneously connecting to wireless Ethernet using Wi-Fi(TM) (802.11b). The simultaneous operation is enabled by Mobilian's interference mitigation and multi-radio coexistence IP.
Expertise to Create and Deliver
Mobilian has almost 100 employees with locations in Oregon, California and Israel. Mobilian's system-level design approach and core competencies in analog design, digital design, and software development allow it to develop and deliver superior multi-standard, simultaneously operating (Sim-OPTM) wireless system solutions. Mobilian's impressive IP portfolio is key to achieving the TrueConnectivity(TM) vision.
The team is comprised of leading technologists from Intel and Qualcomm, blending wireless communications innovations with expertise in high volume Integrated Circuit design and manufacturing. Proven business expertise from Dell, Intel, and SGI, among others, further strengthens the team.
About Mobilian Corporation
Founded in February 1999, Mobilian is a wireless systems company that creates ASICs, IP Cores and software which enable computing and communication devices to support multiple, simultaneously operating wireless radio standards. Mobilian's initial products are focused on the Wireless Personal Area Networking (WPAN) and Wireless Local Area Networking (WLAN) markets. For more information go to the Mobilian web site at www.mobilian.com.
Legal Notices Disclaimer: The information in this press release is subject to change by Mobilian Corporation ("Mobilian") without notice.
Trademarks Mobilian(TM), TrueRadio(TM), TrueConnectivity(TM), Sim-OP(TM) and Coexistence without Compromise(TM) are the property of Mobilian Corporation. All other trademarks are provided for informational purposes only, and are the property of their respective owners. The information provided in this press release is provided "as is."
======================================================
Can Bluetooth Sink Its Teeth into Networking? -- Bluetooth doesn't want to be just for consumers. The short-range radio standard means business, but it could be seriously disruptive to your network.
Network Magazine, Nov 1, 2000 p54
By Dornan, Andy
Network nodes used to be fairly well-defined: mostly desktop computers, the occasional server, and perhaps a more exotic device such as a firewall. Not anymore. The Bluetooth standard aims to make networking ubiquitous, fitting cheap, single-chip radios to literally everything. If its proponents are to be believed, you'll eventually have a network connection in your wristwatch, the garbage can, the office photocopier, and even your coffee cup.
Then again, the initial hype said Bluetooth would be here by the beginning of 2000. Almost a year after that deadline, the industry is still struggling with interoperability issues and expensive chips. Products have been announced, even demonstrated, but nobody is actually using them except in vendor labs and tradeshow halls.
When Bluetooth does finally arrive in mid-2001, it won't necessarily be the seamless networking solution that its promoters hope for. Some analysts even claim that the technology simply won't work. "Using Bluetooth to connect diverse devices from multiple manufacturers is doomed to fail," says Galen Schreck of Forrester Research (www.forrester.com).
He may be right, but network managers still need to be aware of Bluetooth's benefits and dangers. On the plus side, it should at least partially eliminate the spaghetti of wires that clutter most offices. On the minus side, it has a hearty appetite for radio spectrum, gobbling up the bandwidth of other wireless LANs.
Network architects might try to ignore Bluetooth, but most will have to deal with it eventually. Though often hyped as a technology for consumer gadgets, it will infiltrate the enterprise, too. Three years ago, many IT departments thought that they could avoid supporting the Palm Pilot and other PDAs, but users simply bought their own and began bringing them into work. In the next three years, the same will happen with Bluetooth.
TRUE BLUE
Bluetooth dates back to a 1994 project within mobile phone company Ericsson. It was originally intended as a way for phones to communicate with accessories such as wireless headsets, the theory being that customers would find this more convenient than sticking a cable into their ears. Though they didn't know it at the time, such a system would also help to assuage recent health scares. Tests by the U.K. Consumers' Association (www.which.co.uk) claim that far from protecting users, wired headsets actually channel radiation directly into the brain.
Ericsson realized that such as a wireless link would need to be very small and cheap, with all necessary equipment fitted to just one chip. It could also be used for much wider applications than linking mobile phones to headsets, requiring multivendor support. In May 1998, the company joined archrival Nokia and heavyweight vendors Intel, Toshiba, and IBM to form the Bluetooth Special Interest Group (SIG, www.bluetooth.com). The name is taken from Harald Bluetooth, a Viking crusader who unified Denmark and Norway in the 10th century. SIG promised that within eighteen months, Bluetooth would do the same to hitherto separate technologies.
If nothing else, Bluetooth demonstrates that vendors can work together. The SIG has grown into an alliance of nearly 2,000 companies, and new startups are joining every day. The rules are simple: Any company can join, but it must grant all other members a royalty-free license to use any Bluetooth technology it develops. This is to prevent the wrangling over patents that has plagued other multivendor efforts, such as Wireless Application Protocol (WAP).
SIG members include almost everyone involved in the computer and networking industries, and plenty of others such as kitchen appliance manufacturers. These manufacturers delight in issuing press releases about how Bluetooth will enable them to embrace networking, leading to criticism that the standard aims more at the much-maligned Internet toaster than at serious business equipment.
Some SIG members certainly encourage this view. They foresee a future where beer cans communicate with your refrigerator and your wristwatch, telling you when they're cold enough to drink. When you throw away the empties, your Bluetooth-equipped trashcan will tell the garbage truck how they can be recycled.
Meanwhile, the fridge could order more beer from an online liquor store, and instruct your car not to start until you've sobered up.
Fantasy? Maybe at first, but Moore's Law tells us that the cost of producing individual chips is in continuous decline. At present, Bluetooth chips cost around $30 each to make, but prices should reach the SIG's target of $4 each within two years. After that, they will continue getting cheaper. Smart cards have already made disposable silicon a reality, while in the past ten years CDs went from an expensive luxury to something that AOL could bombard us with by the million. According to the Media Lab at the Massachusetts Institute of Technology (MIT), by 2010 each one of us will personally own 5,000 Internet-connected products.
Though Bluetooth may ultimately connect everything, its initial ambition is to be built into every home and office appliance. "Your photocopier will know if it's broken, or if it's about to break," says Tony Valentino, Bluetooth business development manager at Alcatel. "It communicates via Bluetooth to a PC, which e-mails the maker. Then the service man arrives with the right part in hand."
THE PIPES OF PAN
Bluetooth isn't just designed for automatic communication between machines. To emphasize its difference from traditional wireless LAN technologies, the SIG describes it as designed for Personal Area Networks (PANs). These will surround each user, embracing every device that they carry. A cell phone or laptop will be able to communicate with the refrigerator at home, the security system in the office, and devices belonging to other people that its owner meets.
Along with PAN, the other Bluetooth buzzword is piconet, which is a tiny network consisting of eight nodes or less. This might seem rather restrictive, but several piconets can be linked together to form a scatternet (see Figure 1) whose scope is unlimited. A piconet is not the same as a PAN, because the vendors hope that many people will eventually carry more than eight Bluetooth gadgets. Not all of these will need to talk to each other, meaning they can be on separate piconets. For example, a headset will normally be linked to only one other device at a time, as someone is unlikely to be listening to music while making a phone call.
Another difference between a PAN and a wireless LAN is the former's low data rate. Bluetooth has a maximum data capacity of only 1Mbit/sec, which translates to a throughput of only 780Kbits/sec once protocol overhead is taken into account. This is not duplex, and it is shared between an entire piconet. However, many piconets can run across the same airwaves. In an environment with no other interference, throughput is reduced by only 1.3 percent for each extra piconet.
Bluetooth's 780Kbit/sec throughput can be divided in a number of ways. Its central protocol, Logical Link Control and Adaptation Protocol (L2CAP), supports two types of data transmission. Packet-switched Asynchronous Connectionless Links (ACLs) are based on Ethernet, and allow asymmetric data rates of up to 722Kbits/sec in one direction. Circuit-switched Synchronous Connection Oriented links (SCOs) are based on Global System for Mobile Communications (GSM) and provide duplex 128Kbit/sec channels. A single piconet can combine the two types, or dedicate the entire bandwidth to one type. If it chooses to use only SCO links, three separate channels are available.
The range of Bluetooth, and hence the size of a piconet, is only 10 meters (33 feet). There is a plan to extend this to 100 meters, but few vendors are embracing it because a shorter range saves power and prevents interference. This is important because Bluetooth operates in the crowded 2.4GHz Industrial, Scientific, and Medical (ISM) radio waveband, which is not really designed for communications.
The 2.4GHz band is theoretically available worldwide, without a license. Microwave ovens and other heaters need to work at around 2450MHz, the resonant frequency of water, so the 50MHz on either side of this is supposed to be kept free. However, with spectrum such a scarce resource, various governments have nibbled away at the band, licensing it for other purposes.
So that Bluetooth will work worldwide, most regulators agreed to adjust their licensing plans so that it could use the frequencies 2400MHz to 2483.5MHz. France and Spain are exceptions-both have a much narrower band available. Bluetooth will still work in these two countries, but it will be illegal because it risks jamming licensed systems.
The industry has developed a special version of Bluetooth which uses the narrower frequencies, so that French and Spanish users are not denied all of its benefits. However, la dent bleue will be incompatible with the world standard, causing particular problems for Europeans accustomed to having their phones work continentwide.
BANDWIDTH BANDIT
Even where Bluetooth doesn't interfere with licensed spectrum, it can still cause problems for other technologies. It isn't the first communications system to avoid licensing by using the ISM band, and as a result this spectrum has become very crowded. Some cordless phones, telco local loop systems and, of particular interest to networkers, most wireless LANs use the ISM band.
Any system that uses the ISM band must expect interference, thanks to all the radiation emitted by medical scanners, heaters, and the like. Bluetooth is designed to be particularly resilient to interference, because it is supposed to work everywhere. Unlike wireless LANs, which are usually installed in relatively "clean" offices or warehouses, Bluetooth chips will eventually be attached directly to interference sources such as microwave ovens.
"Bluetooth is what we call a rude radio," says Aaron Bennet, vice president of marketing at wireless LAN vendor Breezecom (www.breezecom.com). "It doesn't check if someone else is already using a frequency-it just starts broadcasting." Surprisingly, Bluetooth enthusiasts agree. "It likes to hog the spectrum," Alcatel's Valentino concedes. "When it turns on, it doesn't care what else is out there."
Like many wireless systems, Bluetooth is based on a technology called frequency hopping. As the name suggests, it rapidly hops between different frequencies. The theory is that if there is interference on one, another will be clear. Bluetooth divides the ISM band into 79 channels, switching between them in a pseudo random pattern 1,600 times every second. This is extremely fast compared to other frequency hopping systems, which gives Bluetooth a decisive advantage.
In a frequency hopping system, any hop that finds interference discards the data packet sent during that hop. This means that if one of Bluetooth's 79 channels is being used by another technology, or another Bluetooth picolan, it will lose only one packet out of every 79. However, if the other technology is switching frequency at only 50 times per second, it will encounter interference from Bluetooth on every hop and so lose every packet.
The popular IEEE 802.11b wireless LAN standard doesn't hop at all; instead, it sends data on several frequencies at the same time in the hope that at least one copy will get through (see "Look Ma, No Cables" page 42). This makes 802.11b slightly more resistant to Bluetooth than a slow hopping system, but an 802.11b transceiver still suffers from poor performance when a Bluetooth chip is nearby. This is quite a likely prospect, as 802.11b is designed for relatively long-range, high-speed links compared to Bluetooth. Vendors often stress that the technologies are not competing except at the Physical layer, envisaging laptop computers that include both.
In tests by wireless vendor Mobilian (www.mobilian.com), 802.11b performance suffered degradation from Bluetoosth unless two 802.11 transceivers were situated within 3 meters of each other, at which point their signals were loud enough to not be affected. Given that 802.11 is supposed to have a range of 100 meters, this is a problem (see Figure 2, page 55). Mobilian reports that, as a result of the interference, some companies that rely on 802.11 have already instituted policies banning Bluetooth. Such policies are easy to implement now, but the practice may become impossible if Bluetooth penetration reaches the levels predicted by the SIG.
Mobilian has a proposed solution called TrueRadio, which features an integrated Bluetooth and 802.11 chipset. It claims that this will coordinate the transmissions, so that interference is kept to a minimum and both systems can achieve their full potential. Several other companies are attempting similar solutions, combining Bluetooth with either other variants of 802.11 or HomeRF, yet another standard contending for scarce ISM spectrum.
RED LIGHT FOR DATA
Competition for radio spectrum isn't the only problem that faces Bluetooth. It is also haunted by the specter of the Infrared Data Association (IrDA), a previous attempt at the same market. Indeed, given the current hype surrounding all things wireless and optical, it seems incongruous that this wireless optical networking system built into almost every laptop PC is routinely ignored.
"A lot of the IrDA ports on laptops don't work," says David Hofacker, North American mobile connectivity manager at Extended Systems (www.extendsys.com), a developer of protocol stacks for Bluetooth and IrDA. "The port is there so that manufacturers can check the 'infrared' box in their advertisements, but it doesn't have the right drivers."
So what went wrong? Most in the Bluetooth industry think that IrDA was ill-conceived from the start, based as it was on relatively vague guidelines rather than strict standards. Interoperability was so hard to achieve that many vendors gave up trying. This in turn put users off, meaning that even when it does work, people will assume that it doesn't. The only way to find out if your laptop supports IrDA is through trial and error.
"Bluetooth took lessons from IrDA," says Hofacker. "We're making sure the same mistakes aren't made." While IrDA specified only a physical transport and hoped that application developers would make it work, the Bluetooth standard includes rigidly defined applications. Its protocol stack borrows some proven technology from other protocols, including IP and, surprisingly, IrDA (see Figure 3, page 56). In order to claim Bluetooth compatibility, a device must implement five core protocols and be tested to work with at least one application.
Bluetooth's application support is contained within standards called profiles, each of which governs a specific function. (For a description of the profiles, see "Lucky Thirteen," page 58.) For example, the ninth profile mandates how Bluetooth carries IP traffic. Before a vendor can claim support for this, it must thoroughly test its device with those of other vendors that support the same profile. Only when it passes all these interoperability tests will it be allowed to carry the Bluetooth logo.
The tests are carried out at Unplugfest, a regular meeting where competing vendors set aside their differences and try to get their prototype gadgets to work. "We sign NDAs, lock ourselves in a room, and test devices for a week," says Hofacker. "They're pretty cool events." The results regarding individual designs are kept tightly guarded, but it's an open secret that interoperability problems are holding back the industry.
Forrester's Schreck predicts that interoperability will remain a problem for the foreseeable future. "Nokia's cell phones will still struggle to talk to a Compaq Pocket PC, even if the wireless connectivity works perfectly." They may all be able to hear each other, but they'll still be speaking different languages.
Naturally, SIG members take a more optimistic view. "Bluetooth is promoted by consumer-type companies," Hofacker says. "These are people long focused on making something easy to use." This is one of the reasons why Bluetooth's progress has been slower than originally hoped: Vendors don't want to launch anything until they're sure it will work.
Ironically, Bluetooth may spur IrDA into becoming a serious competitor. "Bluetooth has driven an interest in short-range wireless technologies," says Hofacker. "And IrDA is always going to be faster and cheaper." While Bluetooth aims for a price of only a few dollars, infrared transceivers already cost only a few cents. The interoperability efforts aimed at improving Bluetooth may indirectly help infrared, which can only be good news for users.
IrDA currently supports various data rates between 115Kbits/sec and 16Mbits/sec-again, trial and error is the only way to find out how fast a given device will go-and the Association believes it can be pushed up to 50Mbits/sec. If it can be made to work, this could give it an advantage over Bluetooth in applications that involve large file transfers. Examples include beaming photographs out of a digital camera, loading music into an MP3 player, or, more useful in a business network, printing long documents.
BLUE CHIPS
Interoperability isn't the only thing delaying Bluetooth. Building a single-chip radio is difficult, and mass-producing one at an acceptable cost is even more difficult. And though virtually everyone is now committed to Bluetooth, this wasn't always the case.
Microsoft was very late to join the Bluetooth SIG, and still doesn't support the technology in any of its operating systems. No present version of Windows works with Bluetooth, and Microsoft says it has no plans to include it in future service packs for Windows 2000 or Windows Me. Its first desktop operating system to have Bluetooth built in will be Whistler, a system aimed at both businesses and consumers, due to be released as Windows 2002.
"It's more complex than the industry first thought," admits Alcatel's Valentino. However, he believes that all of the problems can be overcome. Bluetooth products will be available in volume next year, though the SIG's prediction of 100 million units shipped by the end of 2001 looks unrealistic. The only products shipping now are add-ons such as CF+ or PCMCIA cards, but some laptops and PDAs should have it built-in within six months.
With so many vendors staking so much on Bluetooth, it can't afford to fail. If it did, it would be an even bigger embarrassment to the wireless industry than Iridium. It might not revolutionize the world, as originally claimed, but it will be a useful networking technology.
Andy Dornan, senior editor, is the author of The Essential Guide to Wireless Communications, published by Prentice Hall. He can be reached at adornan@cmp.com.
Resources
Aside from magazines like ours, and the vendors themselves, there's really only one site worth listing as a source of information about Bluetooth. At www.bluetooth.com, the Bluetooth Special Interest Group (SIG) has all the official specifications and other data available for free download. It also promises a comprehensive directory, with photographs, of every Bluetooth product actually made. Despite the involvement of around 2,000 vendors, this directory is still rather small.
---
Lucky Thirteen
Bluetooth's Application layer actually consists of nine separate standards. Each one is known as a profile, and includes the protocols and programming necessary for a particular application. Not every Bluetooth device will support all nine applications, and some may support only one.
This system has been criticized as over-complicated, particularly as future versions of the technology will need to add more profiles for new applications, but the Bluetooth Special Interest Group (SIG) says that strictly-defined standards are necessary to ensure compatibility. The profiles are modular, meaning that many can reuse the same code. For this reason, there are an extra four profiles that don't correspond to specific applications but instead are used by others as building blocks. In the list below, these four are numbers 1, 2, 5, and 10.
1. Generic Access Profile. This is the core Bluetooth profile, responsible for maintaining links between devices, as well as for establishing which other profiles are available and for security. All Bluetooth devices need to include this profile, but it in itself is not sufficient for any useful applications. It includes functions necessary to use all of the core Bluetooth protocols.
2. Service Discovery Application Profile. This profile lets a user access the Service Discovery Protocol (SDP) directly, to find out which Bluetooth services are available from a given device. SDP is included as part of the core, but without this extra profile it can only be accessed by applications, not by the user directly.
3. Cordless Telephony Profile. This is designed for what the SIG calls a 3-in-1 phone, meaning a cell phone with a Bluetooth chip that enables it to be used as a cordless phone. It connects to the PSTN via a Bluetooth access point in the home, office, or even a public space such as an airport, preserving cellular capacity and reducing bills. It runs over the Telephony Control Service (TCS) protocol.
4. Intercom Profile. Also based on the TCS protocol, this profile supports two-way voice communication between Bluetooth devices. It is simpler than the cordless telephony profile in that it is only designed for devices within range of each other, not calls routed over the PSTN or the Internet.
5. Serial Port Profile. This profile lets Bluetooth devices emulate a PC's serial port, using the RFComm protocol. It can emulate either an older RS-232 or a newer Universal Serial Bus (USB) cable, and is used by many higher-level profiles.
6. Headset Profile. This profile specifies how Bluetooth can provide a wireless connection to a headset containing earphones and, potentially, a microphone. Intended to be used with computers, MP3 players, and other devices, in addition to mobile phones, it is the only profile so far which uses the AT commands originally designed to control modems.
7. Dial-up Networking Profile. This profile is designed for computers connecting to the Internet via a cell phone. At present they have to do this through a special cable or, perhaps, IrDA, but in the future they should be able to use Bluetooth. It includes the serial port profile and PPP , used by ordinary modems carrying IP over a phone line.
8. Fax Profile. This is very similar to the dial-up networking profile. It enables a mobile phone to emulate a fax modem when connected via Bluetooth to a laptop PC with fax software. Like the dial-up networking protocol, it uses PPP and the serial port profile.
9. LAN Access Profile. This is intended for IP data networking, enabling PCs or smart phones to form a small wireless intranet among themselves. It is also used by access points for connection to fixed networks, either LANs or the Internet. Though most access points will be based on Ethernet, they all act as IP routers-and so can theoretically be built for other layer-2 standards such as Token Ring or SONET. Most experts believe that this will be the most useful Bluetooth profile, eventually used for Voice over IP and Internet-equipped appliances.
10. Generic Object Exchange Profile. This profile controls how Bluetooth uses Object Exchange (OBEX), a client-server protocol taken from IrDA. It allows applications to exchange data directly, without having to use IP , and forms the basis of profiles 11 through 13.
11. Object Push Profile. This profile governs the exchange of electronic business cards in the vCard format, which will be familiar to many PC and Palm users as .vcf files. They contain the same information as regular business cards, but can automatically be filed into a Personal Information Manager (PIM) or a database. The "push" in the name refers to the fact this information must actively be given out by a user; to protect privacy, electronic business cards are usually not just beamed out to every device in range. They can, of course, be refused, or exchanged between two users in a single action.
12. File Transfer Profile. This allows a device to access files stored on another device, in a manner similar to FTP . Apart from linking two computers together, applications for this profile might include printing and scanning.
13. Synchronization Profile. This is designed to keep the data stored on different devices up-to-date. Synchronization can be automated, so that a computer will automatically sychronize data with a mobile phone or PDA whenever they are within range of each other. Most devices accessing this profile will use SyncML, a markup language designed for transferring data between otherwise incompatible PIMs and organizers.
networkmagazine.com |
