Here's some info on the Bluetooth wireless communications standard:
The Bluetooth Standard
By Jeremy Hendy
Communications Systems Design
Legend has it that the Danish King, Harald Blatand "Bluetooth" II, vanquished many a foe in order to unify 10th-century Denmark and Norway. However, the path to the Bluetooth standard appears considerably smoother a thousand years later.
Today's Bluetooth is one of the most talked about wireless standards in the world. Bluetooth was hatched as a short-range, wireless-con-nectivity standard last year by Ericsson, IBM, Intel, Nokia, and Toshiba. The group's stated goal was to "revolutionize wireless connectivity for personal and business mobile devices."
Bluetooth is an open specification that companies can adopt on a royalry-free basis. The standard calls for a short-range radio that enables information exchange, permitting wireless connections between mo-bile phones, PCs, personal digital assistants (PDAs), digital cameras, and other portable information devices. Bluetooth's frequency is on the free, universal 2.45-GHz industrial, scientific, and medical (ISM) band, so the standard is universally portable and interoperable.
Currently, there are more than 500 companies supporting the Blue-tooth standard, and end users are clamoring for more. Why? Simply put, the Bluetooth standard solves some of the most common and vexing problems with portable computing and communications.
No longer will portable computing "road warriors" be forced to pack yards of disparate cables to link their devices. As more devices adopt Blue-tooth technology, portables will seamlessly interface with printers, fax machines, other PCs (in an ad hoc net-work), and communications devices like cell phones. Since the standard supports pretty healthy data rates, Bluetooth is capable of transferring both voice and data.
What's really turning the industry on though - aside from the huge demand for the painless connectivity provided by Bluetooth - is the rela-tively low cost required to integrate this type of device. Initial bill-of- materials costs are expected to be around $20 to $30, but these prices should drop rapidly as the technolo-gy becomes more of a commodity.
As Bluetooth becomes integrated into devices such as PCs, digital cam-eras, cellular phones, and wireless headsets, the diversity of applications will lead to a new set of design challenges for companies whose only previous experience with RF has been in suppressing unwanted emissions. Several companies are helping these manufacturers add wireless capability to their products by offering develop-er's kits, core technology, and design services to accelerate Bluetooth-enabled products to market.
This article details ten common Bluetooth concerns and questions.
1. How low Is "low cost?"
The Bluetooth specification has been written specifically to allow low-cost, highly-integrated imple-mentations. For example, the mini-mum radio receiver sensitivity is a relaxed -70 dBm, and the nominal transmit output power is specified as 0 dBm (I mW), thus removing the need for an off-chip power amplifier.
The long-term target is a bill-of-materials cost of less than $5 per Bluetooth endpoint, which is realistically achievable with a sin-gle-chip, CMOS-integrated RF and baseband solution.
2. What's the range?
With nominal 0-dBm transmit power, typical radio range will be from 10 cm to 10 m. It can be extended to 100 m by increasing the transmit power to a maximum of 20 dBm. For higher-power devices, power control can opti-mize the output power, depending on the range, for minimum interfer-ence and power consumption.
3. What does Bluetooth look like?
Bluetooth devices form small, ad hoc networks, referred to as piconets. A piconet can have between two and eight devices actively communicat-ing with each other, but additional de-vices can be "parked" and accessed when required. Within a piconet, one Bluetooth device acts as the "mas-ter," determining the frequency-hop-ping pattern, packet timing, and coordinating transmissions to the other "slave" devices.
Slave devices can also be mem-bers of more than one piconet at a time, forming an ad hoc "scatternet" of multiple piconets.
4. Is Bluetooth circuit -or packet-switched?
Bluetooth supports both circuit- switched and packet-switched con-nections. The circuit-switched synchronous connection-oriented (SCO) protocols reserve timeslots for time division multiple access (TDMA)-like transmission of voice at 64 kbps. The packet-switched asynchronous connectionless (ACL) protocol can concatenate the remaining Bluetooth timeslots to achieve a maximum asymmetric data rate of 721 kbps (with 57.6 kbps in the reverse direc-tion), or a symmetric 432.6-kbps link.
5. How does Bluetooth avoid interference?
The 2.4-GHz ISM band is shared between many different interfering systems, and a particular problem for voice transmissions (where retrans-mission is impractical) is interference from microwave ovens. Bluetooth mit-igates against interference by employ-ing a relatively fast frequency-hop rate of 1600 Hz. Forward error correction (FEC) and the use of a voice-coding scheme that degrades gracefully in the presence of bit errors, provide additional tolerance to interference.
6. Is it secure?
Bluetooth uses software authentication to create a database of other "trusted" devices. The user will typically activate a registration pro-cedure on both Bluetooth devices, and enter a short PIN number on each before the two Bluetooth de-vices can communicate.
Encryption, also included on the air interface, can provide protec-tion against eavesdropping by using a key derived from the authentica-tion procedure.
7. What modulation scheme is used?
The Bluetooth air interface employs a 1-Mbps Gaussian-filtered frequen-cy shift keying (GFSK) modulation scheme. This allows a straightfor-ward radio implementation, typically using low-cost direct voltage con-trolled oscillator (VCO) modulation and a limiter/discriminator receiver architecture.
8. What voice-coding scheme Is used?
Bluetooth employs a low-cost, 64-kbps continuously variable slope delta (CVSD) modulation voice-coding scheme. This scheme is more toler-ant of bit errors than comparable voice-coding algorithms such as adaptive differential pulse code mod-ulation (ADPCM), with errors show-ing up as an increased noise floor rather than "pops" or "clicks." A 64-kbps p-law or A-law pulse code mod-ulation (PCM) may also be transmit-ted across the air interface, when transparent connection to digital voice networks is a requirement.
For higher quality voice connec-tions, the TDMA characteristics of the SCO link are changed to provide a higher-bandwidth physical layer connection. FEC is then used to reduce the error rate.
9. What about type approval?
One of the main objectives of Bluetooth is to enable simple, "out-of-the-box" operation of Bluetooth products by the end user. Bluetooth standard-ization committees are creating a comprehensive set of basic compli-ance and interoperability tests. All Bluetooth products will be required to pass the basic Bluetooth compliance tests, the relevant interoper-ability tests, and local government type approval testing (FCC part 15). Products completing these tests will be Bluetooth certified, and will dis-play the Bluetooth logo.
10. When can I expect to see Bluetooth products?
There are already some preliminary components available that imple-ment a subset of the Bluetooth func-tionality. However, fully Bluetooth- compliant components aren't expect-ed to be in production until late into the third quarter of 1999, with the first products likely to be launched towards the end of 1999. Expect to see a whole wave of Bluetooth prod-ucts launched in the first half of 2000.
Jeremy Heady is a business director at Cadence Symbionics design center based in Cambridge, England. He has a degree in electrical engineering from the Univenity of Liverpool. He can be reached at jphI@symbionics.co.uk.
|
