Thought Control
By Mike Grebb, Mon Dec 02 12:30:00 GMT 2002

The Interuniversity MicroElectronics Center in Belgium is developing wearable technologies that'll read your mind, and then some.


By now, efforts by wearable computing enthusiasts to outfit human beings with semi-attached gadgetry has failed to elicit much interest beyond Star Trek fans seeking assimilation by the dreaded Borg. But rumors of the death of wearable computing aren't just exaggerated; it's turning out that they could be dead wrong. Just ask any of the researchers at Belgium-based Interuniversity MicroElectronics Center (IMEC), which foresees a world in which the marriage of wearable computing with wireless capability - combined with a focus on immediate markets such as home healthcare and athletic training - could allow each individual to become his or her own wireless body-area network (WBAN).

Baby Boomers Are Getting Old


Imagine a world in which people can each exist inside a small wireless bubble, with sensors close to the body constantly sending information on such functions as heart rate, blood glucose levels, or whatever data might be needed for a particular medical condition. The sensors could communicate wirelessly with a small, low-power device sewn into clothing or attached to a belt. And imagine that those small devices are multi-modal and therefore able to seamlessly transmit data back and forth - perhaps to a hospital or family member - over a broader terrestrial or satellite wireless network.

"The time is about right to start a more integrated approach to these kinds of technologies," says Bert Gyselinckx, director of IMEC's Human++ program, which is working on melding wireless, microprocessor and power supply technologies into cheap, low-power devices. "We want to bring all of these aspects together," he says. "There's a large need for these type of sensing networks in the medical world. The baby boomers are getting old. All of these people would like to get old at home."

Eventually, experts say such devices could filter down into the consumer population, especially for family members faced with the difficult choices surrounding elderly care for all of those baby boomers. Take a grandmother who wants independence but needs to be monitored all the time. "It be very nice to be able to ring her up without bothering with the telephone, and get a feel for whether she's moving around or is okay," says Dennis Silage, professor electrical and computer engineering at Temple University in Philadelphia. "It's a brave new world out there."

Gyselinckx notes that the WBAN concept really began with the first pacemakers decades ago, but that the convergence of new technologies in recent years has opened up new possibilities. For example, IMEC has been working on a wearable device that could deliver electrical signals into the body to suppress pain. It could be worn and controlled via a wireless connection. That way, recovering patients could continue to receive powerful pain suppression at home, cutting down on hospital stays. And physicians could even monitor the devices from the hospital via a broader wireless connection.

IMEC researchers are also trying to build a wearable electroencephalograph (EEG) machine, which can measure the electrical activity of the brain and transfer the information gathered to a report. Such machines are now used by hospitals to monitor people with brain disorders such as epilepsy. "They put you in a hospital with a bunch of electrodes on your head and bunch of cables, and you can't move at all," Gyselinckx says. A wearable EEG machine, on the other hand, would enable a patient to move freely and self-monitor from home. Another wireless connection could send an EEG report to the hospital.

"I Think You'd Better Lay Down"


In the future, Gyselinckx says such systems could constantly run brain scans during the day and send wireless alerts to a personal digital assistant (PDA) or similar device warning the patient that a seizure was imminent. That way, the patient could go lay down or get away from sharp objects to avoid injury during the seizure. The next step would be devices that automatically intervene to stop the seizure by perhaps signaling an affixed or surgically implanted device to inject medicine at the exact right time.

At the same time, IMEC runs another division, M4, which seeks to combine all of the necessary multi-modal components into wearable devices so that WBANs can work seamlessly, allowing them to communicate via short-range signals with sensors on the body as well as access other longer-range wireless networks. It's not an easy feat. The most straightforward way to build such a device is to simply include a separate radio receiver chip for every frequency used. But each new chip added increases the weight, size and cost of a device. "There's no way to do this with a single-chip solution," says Liesbet Van der perre, director of IMEC's M4 program.

The solution, Van der perre says, is a "system-in-a-package" concept in which a device could receive several different frequencies with a reconfigurable digital system and software-defined radio. So the Human++ and M4 programs have started working more closely together in an effort to create devices that are tightly integrated.

In fact, IMEC envisions a "global system for ambient intelligence" (GSAI) that would replace the existing Global System for Mobile Communications (GSM)-based wireless systems now prevalent throughout most of the world. The result would be a "wearable assistant" that could integrate global frequencies in one device offering everything from multimedia games to biometric monitoring and potentially even location-based services using global positioning satellite (GPS) systems.

Will We Be Assimilated?


Of course, several hurdles stand between such devices making their way from laboratories to the marketplace. Even after researchers work out the technology issues and develop the small, low-power gadgets they envision, it's unclear how widespread they will become and how fast. Gyselinckx admits that it's unlikely such devices will become remotely common until after 2008. "It will gradually become more and more widespread," he says. "But to be ubiquitous, we need a couple of technological breakthroughs." Namely, he says that researchers still must conquer the trade-off between computation and power, including simple power storage issues within the devices themselves. "Batteries, unfortunately, are not evolving rapidly," he says.

Van der perre agrees that power-consumption issues will continue to be among the biggest challenges for WBAN researchers, who are now trying to perfect microelectromechanical systems (MEMS), and voltage and clock scaling technologies to lower the power requirements. "The consumer will always want to have more battery time and more features," she said. "The sky's the limit there."

Because battery power technology is progressing slower than Moore's Law, IMEC has focused on lowering power consumption of devices using MEMS and other novel approaches such as solar cells and by using body heat and movement as supplemental power sources. Although such natural energy sources can't power a device on their own, they can increase battery life considerably. "It's an added value, whatever you can get from the outside," she says. "But the battery will not go away."

The bottom line, however, is that researchers must find a way to lower power consumption while continuing to provide the processing speed consumers expect. In fact, Moore's Law has conditioned consumers to expect a doubling of speeds every 18 months, a trend unlikely to change any time soon. Jim Tully, a Gartner Dataquest analyst who has followed IMEC's activities over the years, say processors could theoretically reach clock speeds of 12 gigahertz within the next decade, presenting WBAN researchers with constant problems as they try to keep devices small and low power.

"Chances are that that processor would run so hot, it would burn a hole in the chip," he says. "That means you have to come up with new architectures." Tully says IMEC may be among the best-positioned research consortiums to solve such problems. "The stuff they do is very, very good," he says. "There's a lot of synergy and reuse of knowledge."

In addition, IMEC has a lot of help around the world. Several universities and research groups are working on similar projects to create wearable devices using sensor technology. Earlier this year, the University of California in Los Angeles (UCLA) received a $2 million grant from the National Science Foundation to build medical "jackets" able to use sensors wired throughout the fabric to detect heart-rate and even administer drugs intravenously at the appropriate time based on the readings.

But while UCLA researchers plan to integrate wireless technology, they are limiting initial prototypes to wired connections. "If there was interference, you could inject a drug at the wrong time," says Majid Sarrafzadeh, professor of computer science at UCLA. In addition, wireless connections back to a hospital could raise privacy and security issues. "You wouldn't want you're insurance company to know your heart rate and such things," he says.

Although IMEC researchers say they have also considered such issues, they are primarily leaving them to the companies who will eventually license the technology. "The manufacturers will deal with these kind of problems," says Gyselinckx. "People often ask me, `Will a pacemaker be connected to the Internet?' It's definitely a big issue."

Societal Questions Abound


Other more basic questions come to mind: How many people will accept the idea of their blood content or heart rate being connected to a wireless network? And how many would see the benefit for such an intrusion? "If it becomes pervasive, what does it mean to the man on the street?" asks Silage. Researchers at Nippon Telegraph and Telephone and the University of Tokyo recently conducted experiments in which two individuals were able to exchange data from their PDAs just by shaking hands: They're bodies acted as electrical conductors.

In the future, IMEC researchers expect to be able to surgically implant WBAN technology, literally allowing human beings to become biologically connected to global wireless systems. At some point in the future, such connectivity might even evolve into bionic telepathy or other science fiction inspired concepts. No one has to worry about such advances any time soon, but it does raise an interesting question: "What are going to accept as human beings?" says Gyselinckx.

It's a question that future generations may be forced to answer.

Michael Grebb has also written for Wireless Week, Business 2.0, and Wired News. From Washington DC, he covers the impact of mobile technology on modern society.