Wireless Technologies
Exploring a Wireless Future for Medical Electronics
A number of wireless options offer new possibilities for the future of medical electronics. Bluetooth technology takes the lead.
William E. Saltzstein
All
new technologies follow a path from inception that includes hype and eventually
arrives at wide-scale availability and deployment. Bluetooth and other wireless
technologies are no exception. Bluetooth technology has certainly received its
share of hype, as well as its share of critical press and reviews.
This article looks at Bluetooth technology in relation to other wireless technologies available to medical electronics developers. What is the reality, and what are its capabilities and costs? In particular, it explores Bluetooth wireless technology in greater depth to gain an understanding of its place in medical device communications. Wireless interference and coexistence are discussed to provide an understanding of the issues and reality within the industrial, scientific, and medical (ISM) communications band (2.4–2.485 GHz). Finally, Bluetooth-based products are discussed with the goal of understanding the issues as well as the status of product development.
Although Bluetooth technology is certainly not the universal solution to all wireless needs, it addresses many of the needs in medical applications. It is a particularly good fit in models that demand high mobility, long battery life, and minimal infrastructure support. Bluetooth does much more than simply eliminate cables. It provides access to a wide range of standard devices and communications options, and it provides the ability to form small networks. Bluetooth enables additional communications links by giving access to wide-area networking through cellular phone data communications as well as access via traditional Ethernet local-area networks (LANs).
Cables: Good and Bad
When a cabled connection fails in a device, there is a straightforward troubleshooting process: check both connections, then replace the cable or replace the connectors. That's it. It is a well-understood process by users and customer-support personnel. The problem is that cables do fail, and medical-grade cables are expensive to replace. These cables require expensive electrical isolation circuitry and extensive environmental testing for every connector in an enclosure. Some cables connect only to limited types of devices. Others require adapters or software changes to work with certain devices. Finally, patient mobility is limited. It is difficult to imagine a patient walking around with 10 m of cable connected to a device.
Although wireless connections are an obvious solution, they work quite differently than cables. Wireless connections can be affected by metallic partitions or improper antenna orientation. Moreover, the connection doesn't simply stop working when its range is exceeded, but rather, it degrades. Unlike a cable that can be checked manually from end to end for integrity, when degradation occurs in a wireless device, the connection cannot be checked manually.
During hazard analysis and usability testing, the device designer must think about wireless connections differently than cabled implementations. User documentation and customer-support material need to be structured accordingly. To assist in understanding and troubleshooting a wireless connection, user interfaces should include features such as quality-of-service indicators.
Wireless Options: Overview
) |
|
Figure 1. Wireless options
available to medical device
manufacturers. |
IrDA. IrDA refers to the Infrared Device Association that oversees the specifications for technology using optical transmitters and receivers for data transmission. IrDA developed from the same technology used in television remote controls and shares some of the advantages and disadvantages of those devices. Since it was the first widespread cable-replacement technology, it has an important place in the wireless landscape. It is implemented in nearly every laptop computer and palm-sized organizer shipped today.
IrDA has a limited range, with high-speed (16 Mb/sec) transmission requiring separations as short as 1 ft for reliable operation and no farther than 4 ft for its slower rates. It requires that the two devices be in line of sight, and it only works between one pair of devices at a time. IrDA is very low in cost and is relatively simple to design into products. Although its line-of-sight operation and range limitations can be a drawback in some applications, these limitations do provide a secure link that can be relatively immune to eavesdropping or other attempts to intercept data being transferred.
IrDA has been called the “most successful failure in recent computing history.” The main issue has been that getting it to operate between devices can be quite difficult, and application support has been limited. Everybody has it, but virtually nobody uses it. The IrDA group has recently instituted new testing standards, and the interoperability between devices continues to improve because of both the association's and the manufacturers' efforts. However, because of the large installed base of older equipment, interoperability issues will remain. Even with its drawbacks, though, in some applications, IrDA will continue to be the best option because of its low cost, ease of implementation, and security. It is currently implemented in medical devices from a number of manufacturers, and it is being discussed by several medical communications standards bodies.
HomeRF. HomeRF was developed to facilitate data exchange between home appliances and entertainment equipment. It combines some of the attributes of IEEE 802.11 (data) with those of DECT (cordless handset capabilities), allowing both voice and data to be transmitted over a wireless network.
Industry analysts outside of the HomeRF group believe it is
unlikely that this technology will be long-lived (many say it is already dead),
and that it is highly unlikely to be seen outside of the home environment or
in mobile devices. One of Home RF's founders and biggest proponents, Microsoft,
joined the Bluetooth Special Interest Group (SIG) as a promoter. Microsoft has
promised Bluetooth—but not HomeRF—support in the upcoming releases
of its operating systems. Intel, one of HomeRF's founding companies, recently
announced that it would not support the 2.0 standard. The primary reason for
HomeRF's likely short life is the popularity of IEEE 802.11b and the high
interest in Bluetooth. The strength of these two technologies combined squeezes
HomeRF from above and below with higher volumes, lower costs, and greater product
offerings.
IEEE 802.11b. IEEE 802.11 was developed specifically for LAN replacement,
using IEEE standard 802.3 (Ethernet) protocols. IEEE 802.11b is designed for
computing devices contained in a building covered by one or more cells. It offers
structured networks with cellular-like roaming for devices moving between access
points.
There are two distinct types of IEEE 802.11 protocols that use the ISM band shipping. The frequency-hopping (FH) spread-spectrum technology is used in several patient-worn medical devices currently on the market, and the direct-sequence IEEE 802.11b (also called Wi-Fi) is seeing great acceptance in hospital, office, and home computing environments. IEEE 802.11b, which offers up to 11 Mb/sec data rates, is being used in many fixed and mobile applications that require high speed. In terms of power, the cost of IEEE 802.11 implementation is much higher than IrDA and quite a bit higher than Bluetooth (roughly 4–10 times the power). Chip set costs for IEEE 802.11b implementations are also higher than for the current generation of Bluetooth.
In the near future, other variations of IEEE 802.11 will be available, including a higher- frequency, high-speed version (IEEE 802.11a at 54 Mb/sec) and a higher-speed version in the ISM band using new coding techniques (IEEE 802.11g at 54 Mb/sec).
IEEE 802.11 is here to stay because of its wide support and implementation. It is also easy to integrate into existing wired networks and works with standard enterprise computing infrastructures. The technology has some issues based on its power, IT infrastructure, and administration that hamper its viability for use in medical devices. Additional concerns include its susceptibility to interference found in certain areas of the hospital, as well as its potential to interfere with other devices.
DECT. Digital electronic cordless telephony (DECT) is a standard designed primarily for cordless handsets, although its specification does include data capabilities. It is a European standard with a North American implementation in the ISM band called wireless digital cordless telephony (WDCT).
Its ability to provide handsets that roam in large corporate
settings is popular in many European countries. The technology is promoted heavily
in Europe by Siemens (Munich). In North America, WDCT is seeing usage primarily
in 2.4-GHz cordless telephones in the consumer market.
DECT's data capabilities have been largely untapped, and the popularity
of IEEE 802.11b and voice-over-Internet protocol (VOIP) emerging standards will
limit its adoption in its primary corporate markets. Industry experts are quite
pessimistic about its longevity and penetration in the worldwide market.
Wireless Telemetry. Although
not included on the chart in Figure 1 (because it is not an industry standard),
wireless hospital telemetry demands some explanation because of its market position.
The wireless medical telemetry system (WMTS) band was created to move hospital
telemetry systems away from interference created by the new digital television
transmissions. It exists in the United States from 608 to 614 MHz with additional
spectrum to be assigned when military users at 1.394 GHz free up that segment.
Additional
devices operate in the 902–928 MHz band and on the 2.4 GHz ISM band. Neither
of these two ISM bands is exclusive to medical uses. Custom solutions in telemetry
are specific to each manufacturer and begin with the original one-way FM analog
transmissions of waveform. One implementation (General Electric; Waukesha, WI)
uses this band to transmit digital and two-way information to facilitate new
devices and models.
Because most of these implementations are customized, there is little or no interoperability across manufacturers, and design and manufacturing can present major issues. In addition, there is no uniform international support for this band, requiring changes and requalification for countries outside of the United States.
Bluetooth. Bluetooth is the newcomer on the radio-frequency (RF) landscape; it originated as a cable replacement technology primarily targeted at personal, portable computing and communications equipment such as personal digital assistants (PDAs), cell phones, and laptops. The technology allows not only the replacement of simple point-to-point cabling (with minimal impact in cost and power) but also the quick and simple creation of small ad hoc networks of devices, called piconets.
The specification was initially created by Ericsson, IBM, Intel, Nokia, and Toshiba. It was named after the tenth-century Danish Viking king Harald Blatand (a.k.a. Bluetooth) who united the warring factions of Denmark and Norway. (King Harald liked blueberries and didn't brush after his snacks, as one legend has it.) One of Bluetooth's main goals is to create a truly international standard that could be implemented identically worldwide. This goal has been met with very few exceptions, and those exceptions will disappear in 2002 when France, Spain, and Japan open up the frequency band for Bluetooth.
More than 2500 companies have signed up as adopters. The initial group of big-name companies has been renamed the Promoters Group and has been expanded to include 3COM, Lucent, Microsoft, and Motorola. Products incorporating Bluetooth wireless technology are now shipping from several of the promoters, as well as from many other companies. This year, IEEE formally adopted Bluetooth as the technology for its personal-area network (PAN) standard 802.15.1. The basic radio covers 10 m in open air (although current implementations operate well at significantly better range) with an optional power boost to allow 100-m operation. It is designed for moderate speed, with a total bandwidth of 1 Mb/sec, and a theoretical 720 Kb/sec payload, dividing the bandwidth between the devices using data channels and voice channels. It supports as many as eight devices on a piconet.
Bluetooth is based on lessons learned from many other wireless technologies and borrows shamelessly from DECT, IrDA, GSM, and other technologies to make for relatively simple implementation with high levels of functionality and interoperability.
Bluetooth: Implementation and Issues
Perhaps the easiest way to get started with Bluetooth is to use an external adapter to prototype the design. This approach enables a designer to get an understanding of issues associated with a particular implementation. It also provides an opportunity to gain valuable user and marketing feedback. With a prototype up and running, it is possible to navigate the make-versus-buy decisions. Several issues must be considered to formulate a product and project plan.
The $5 Solution. A lot has been said in the press about the $5 cost goal for implementation of Bluetooth wireless technology. It is clear that this goal will be attained in the near future (some suppliers claim they have met this goal), but it is important to understand what this goal means. The industry goal of $5 includes only the RF and baseband signal processing functions. For implementation, Bluetooth requires external components (at least an antenna or board space for an on-board implementation; often other discrete components are necessary as well). RAM, ROM, and processor bandwidth are also required to implement the entire Bluetooth software stack from the baseband layers up through the profiles and application. In addition, all of the current Bluetooth software stacks that implement more than the headset profiles require an operating system.
The $5 goal assumes that a device (such as a cell phone) already has an operating system and enough of the resources listed above. The goal also assumes a royalty-free implementation of the Bluetooth software stack, which is not the case unless a designer wants to implement the rather complex specification. The goal also assumes volumes in the millions of units per year. Unfortunately, few of the conditions mentioned above are met for medical devices, and thus realistic implementations in lower volumes are unlikely to meet that cost goal. For the next year, realistic estimates range from $25 to $60 in low volumes, but this cost estimate would still require varying amounts of development resources to reach those price levels. Complete alternatives that require little or no development cost more, but yield quick time-to-market. Such alternatives may be the best for low-volume production, or until customer acceptance justifies highly integrated solutions.
RF and Antennas. Most medical device manufacturers have extensive experience—usually negative—with RF. A great amount of time and money is usually spent trying to reduce RF emissions and reduce susceptibility. Bluetooth relies on being able to transmit and receive RF signals, which is quite a different skill set from the usual design requirements.
Fortunately, many excellent designs are available for Bluetooth that include all of the RF components, eliminating this part of the design task. However, the best Bluetooth RF transceiver will function poorly unless proper attention is paid to the design and location of the antenna.
The antenna needs to be located properly to limit RF shadowing outside shielded cases and components. It also needs to be designed with the appropriate gain and radiation pattern for a given application and its placement in the device. Standard 2.4 GHz antennas are available that offer many options for placement, pattern, gain, and physical mounting. Board-mount versions are available if the enclosure is radio-transparent and the radiation pattern is appropriate for the application. It is important to note that plastic material selection can make a difference for embedded antennas since these materials vary in their transparency at 2.4 GHz.1
Software Support for Host Stack. Many medical devices do not need the expensive, highly integrated solutions that eliminate the host communications interface (HCI) and merge the baseband and host processing. In most situations, designers will use a module that incorporates the baseband, RF, and a Bluetooth software stack in the host software. Depending on the type and number of profiles supported (i.e., amount of functionality required), a Bluetooth stack will require:
- 32–64 Kbyte of code space.
- 16–32 Kbyte of RAM.
- ~5 MIPS of processor bandwidth.
These numbers are approximate and represent a conglomeration of several suppliers' specifications.
Regulatory Issues
Three approvals are required for medical devices in the United States: Bluetooth, Food and Drug Administration (FDA), and Federal Communications Commission (FCC). In Europe and Asia, other regulatory agencies are responsible for device approvals and telecommunications type approval. Bluetooth approval is new for medical device manufacturers, and devices without wireless components have also been exempt from FCC approvals in most cases. FDA's approval process is always a challenge and presents issues with new technologies.
Bluetooth Qualification. Bluetooth technology has a strict enforcement policy, and the Bluetooth qualification process is quite well defined in SIG documentation. To have free license to the intellectual property that the SIG members have contributed, a product must pass the appropriate qualification process before sale. Once qualified, the product may then display the Bluetooth brand, logos, and labeling.
The qualification process is meant to guarantee to customers and users a high level of interoperability between devices. It is taken quite seriously for Bluetooth products. The process appears to be effective based on a sampling of products that meet the current version 1.1 specification.
The Bluetooth qualification program allows components to be qualified to various levels. Hardware and software components can be qualified so that a device manufacturer could integrate them into a device and not be required to repeat the portion of the testing that the component has already passed. This means that if a medical device manufacturer integrates an entire product-level-qualified OEM module into a device, no additional testing is needed for the device to be qualified for Bluetooth. Other combinations of qualified components have different sets of implications, so it is important to understand what qualification work will be required when evaluating the trade-offs of make-versus-buy.
Requirements. Bluetooth wireless technology is royalty free for Bluetooth SIG members, but there is a hitch: products offered for sale must complete the appropriate testing. The test results must be reviewed and approved by a Bluetooth Qualification Body (BQB). The Bluetooth brand explicitly implies interoperability and can be used only on qualified products, which is meant to be a benefit for manufacturers, integrators, and end customers. If the Bluetooth intellectual property were used without qualification testing, the license would not be in effect and the offending company would be infringing on certain patents.
When product-level devices are used in other products, such as an approved PCMCIA card and its associated Windows software, no additional approvals are required for use in a product to be resold to an end customer. The company integrating the Bluetooth component does not need to join SIG.
FDA. In the highly regulated medical device industry, one of the important factors in technological adoption is the path to approval for that technology when integrated into products and systems. Depending on the implementation and usage of Bluetooth, there are several approaches to approval with FDA and international approval organizations. It is important to note that these are potential paths and must be reviewed and approved by the manufacturer's regulatory affairs staff.
If Bluetooth wireless technology is added to a device as an external adapter with no modifications (such as replacement of an existing RS-232 cable), and the device is approved for data communications capabilities through that port to external equipment, it may be possible to use the approach taken by many device companies with external modems. Documentation from the Bluetooth Qualification Test Facility (BQTF) and BQB provide evidence of compliance verification.
It is clear that if the technology is incorporated internally to a device, a 510(k) application of some type is required, especially because software and hardware modifications are likely involved. The use of approved and well-tested components that have a track record for external use (as in the case above) will be easier to analyze separately. It is also easier to demonstrate to regulatory agencies that such components are safe and reliable. The external approach also allows easier testing and enables device manufacturers to gather and include clinical and customer usage data with a submission.
Currently, only one Bluetooth-based medical device (Ortivus; Täby, Sweden) is being marketed and distributed. Wireless and telemetry devices have been shipping for decades in the United States and worldwide. Potential predicate devices using IEEE 802.11 frequency-hopping technologies have been allowed by FDA, and some of these devices are currently being marketed and distributed.
FCC. FCC testing and approvals are standard practice in the telecommunications industry, but these regulations have not applied to medical devices to date. Once Bluetooth is added, FCC testing becomes mandatory because the device becomes an intentional radiator. Bluetooth RF modules have been designed with FCC compliance as a key parameter, so if they are correctly incorporated into designs, they should be approved. A potential risk to achieving FCC compliance for designs using pretested components comes from internal systems emissions. If these emissions are coupled into the Bluetooth RF and reradiated by the Bluetooth transmitter and antenna, the device may not comply.
Make Versus Buy
The decision about which parts of the technology to purchase and which parts to design in-house presents new issues with Bluetooth wireless technology. Careful evaluation should be made as to the requirements of Bluetooth on the system components in terms of hardware and software resources. Qualification—both Bluetooth and FCC—must be considered in the project plan. The levels of in-house expertise compared with an outside consultant for high-level integrated modules must be evaluated early in the project plan.
It is easy to add up the individual components on a cost-per-part basis and come up with a low-priced solution, but this may produce a disastrous outcome when the actual costs of engineering and qualification are fully understood. A higher-priced, highly integrated solution may save substantial nonrecurring expenses. It also speeds time to market, which often offsets the higher price for an entire product generation. It is critical, therefore, to understand all development costs and the schedule and how these may impact revenue calculations.
RF Issues
Interference and Susceptibility. The ISM band is significantly outside the range of naturally occurring biological signals, and so the risk of interference is most likely to the product components in a particular design. Bluetooth is a very low-power transmitter and unlikely to interfere. Most current device standards, including those for implantable devices, require such testing for susceptibility in the ISM band.
Some concerns have been discussed regarding microwave ovens, which use the ISM frequency band. This issue has been explored by other medical device companies using FH devices and has been found not to be an issue.2 Although the effects on Bluetooth devices are similar to those of IEEE 802.11 FH, interference is less likely because Bluetooth hops faster and operates on significantly lower power levels.
Coexistence in the ISM Band. Because of their experience with hospital telemetry systems, medical device companies and regulatory agencies have a heightened sensitivity to the issues of interference and coexistence. Hospitals may already have installed wireless LANs, and multiple ISM radios may exist in other environments, so these issues are very real.
Testing performed by many companies has shown that although performance degradation occurs (less than 15% when antennas are within 1 ft, and less than 5% in most other cases), the real-world effects of interference are manageable by current applications. To entirely eliminate such issues, more work is necessary, but for now the issues do not seem to be significant in most usage models.
For the future, specification groups (e.g., the Bluetooth SIG and the IEEE 802.15 working group) for all of the wireless technologies are working toward interoperability solutions that would eliminate interference issues entirely. In addition, several companies are already working on products that incorporate both Bluetooth and IEEE 802.11b into a single device. It has been predicted that both technologies will be used simultaneously in several scenarios.
In June 2002, FCC made a ruling allowing ISM band FH technologies to use fewer than 79 bands. This ruling allows a technique called adaptive frequency hopping, which will enable the development of solutions incorporating both IEEE 802.11b and Bluetooth wireless technologies.
Current Products
Original predictions for the adoption of Bluetooth wireless technology were far too optimistic. Although this is true of nearly every new technology, it is nevertheless unfortunate.
) |
|
Figure 2. Application of
Bluetooth and 802.11b technology.
|
Products are being approved to the Bluetooth 1.1 specification and released to production at an increasing rate. These products are more prevalent in Europe, where universal standards for cellular communications and acceptance of all things wireless are higher. But, these products are also being sold by major e-tailers in the United States, and major carriers currently have Bluetooth-enabled handsets available. The development of Bluetooth products in the United States will likely increase dramatically this year. The following sections provide an overview of products that are announced and shipping with Bluetooth wireless technology.
Approved Bluetooth 1.1 Products. Figure 3 shows a graph of the products approved to the current Bluetooth 1.1 specification as of July 23, 2002. It is strongly recommended that only components and products based on the 1.1 specification be used.
) |
|
Figure 3. A graph of the
products approved to the current Bluetooth 1.1 specification.
|
Medical. The first medical device listed on the Bluetooth Web site was an approved product from Ortivus. In addition, several studies have been or are being conducted using Bluetooth as a method for synchronizing personal devices and transferring data in hospitals, particularly in Europe. At the most recent Bluetooth Congress, a panel discussion on the use of Bluetooth in medical applications included a study being performed at the University of Mainz Medical Center (Mainz, Germany) involving Bluetooth use with PDAs and access points.
Cellular. Bluetooth-enabled cellular handsets are currently in production and shipping from Ericsson, Motorola, Nokia, and Samsung. Cellular handsets are now available from AT&T Wireless, VoiceStream, and Cingular, and other cellular carriers are currently qualifying the handsets for their own distribution.
Headsets. Hands-free wireless headsets are one of the original product applications for which Bluetooth was developed. With recent political and legal pressures prompted by a concern for driver safety, this type of product is seeing even greater demand. Headsets from Ericsson and Motorola are shipping, and Plantronics will be shipping soon.
Several other in-car uses for Bluetooth are being developed and have been publicly demonstrated by major automobile and automobile electronics manufacturers. Several companies are shipping hands-free kits for car installation.
All of the major U.S. and European car manufacturers have made public announcements that they intend to have Bluetooth installed in all new cars by the year 2005 for hands-free voice and data connectivity and potentially for other telematic systems within the vehicle.
Personal Digital Assistants. PDAs are also a major focus for Bluetooth wireless technology. Bluetooth products are shipping for the Palm OS and Pocket PC platforms, including the iPAQ 3870 and 3970 with integrated Bluetooth (see photo). Palm has announced Bluetooth support for its OS version 5, and rumors abound that its next generation will have Bluetooth built in. Add-ons in the form of sleeves or sleds and plug-in cards are available for many recent models. Current operating systems require additional software drivers, but both Palm and Microsoft have announced integrated support for Bluetooth in the next revision of their respective operating systems.
Laptops. Laptops incorporating Bluetooth are shipping from major manufacturers (IBM, Sony, Compaq, Hewlett- Packard, and others). Bluetooth PCMCIA cards are currently available from many manufacturers for computers without built-in support.
Software support is provided for major Microsoft operating systems, and several projects are in various stages of completion for supporting Bluetooth under Linux (to date none have passed the Bluetooth qualification process). Apple is also shipping software for Bluetooth on OS X.
Microsoft has announced that its next revision of Windows XP will support Bluetooth with clear support for the new PAN profile, as well as the human interface device (HID) profile.
Peripherals. Printer and human interface profiles have recently been released to the Bluetooth SIG, and the market will soon see many different types of input, output, and storage devices. One of the first completely integrated products is the HP Deskjet 995C color ink-jet printer. Microsoft has announced that it will produce Bluetooth mice and keyboards supporting the HID profile along with the upcoming support of Bluetooth in Windows XP. Another important peripheral is the LAN access point (supporting the Bluetooth LAN access profile). These products provide the functionality to manage a piconet of devices and connect them to an IEEE 802.3 Ethernet LAN. These are currently shipping from several different companies and are the first products to take full advantage of the Bluetooth piconet.
Cards and Dongles. Bluetooth PCMCIA and Compact Flash (CF) cards are available from many different companies and are supplied with software that supports handhelds and laptops. Although the Bluetooth specification is quite exacting to ensure interoperability and compatibility, there is no specification of the user interface (apart from terminology). This has already produced a wide range of implementations.
Several companies offer parallel adapters with built-in printer profiles. It is important to note that some are not yet compatible with the recently released printer profile. Manufacturers vary in their profile support. CF and PCMCIA profiles to look for include dial-up networking, serial port, LAN, point-to-multipoint piconet support, printing, and object exchange. Adapters intended for printers need to support the printer profile.
Bluetooth Adapters. Adapters are currently available that target general-purpose adaptation of existing devices. These adapters relieve the host of the software and hardware burden of the Bluetooth stack. Medical device manufacturers can use adapters to prototype, test, and ship products quickly. In addition, adapters can be used to enable currently installed products with new communications features.
Another unique set of products (connectBlue) allows adaptation of RS-232 devices as well as development of integrated products. Some models include features such as an integrated Web server (see photo). Adapters need to support the profiles required by their intended applications, which most likely will be serial port, dial-up networking, and LAN.
Conclusion
It is clear that medical devices and medical information management products will benefit from the advantages provided by Bluetooth wireless technology. And although this technology certainly presents new challenges, the benefits appear quite significant in both current and future use models for integration into medical products.
References
1. J Oganado, "Plastics with a Bluetooth Bite," Design
News no. 13 (2001): 65–66.
2. Data published in various articles by Welch Allyn Protocol; see Medical
Strategic Planning: Industry Alert 3, no. 1 (2001): 19.
William E. Saltzstein is president and founder of Code Blue Communications, a company based in Redmond, WA, that has been developing Bluetooth technology.
Copyright © 2002 Medical Electronics Manufacturing
