This is the second half of a two-part series on technology breakthroughs that have the potential to change computing. Last week, we looked at five chip-level innovations that will make electronic devices faster, more powerful, more flexible and less expensive to manufacture. This week, we explore advances in how we access the Net, how we power our devices and how we interact with them.
From over-the-air power to neural computer control, each of these technologies has the ability to fundamentally alter the digital landscape. Put them together with the circuit advances we discussed last week, and you get a revolution in the way computers and electronics are designed, manufactured and used.
Extreme wireless: Multi-gigabit Wi-Fi
As great as it is to be able to grab data out of thin air, Wi-Fi is basically 1990s technology that's been jazzed up several times. "We want to take Wi-Fi truly into the 21st century," says Ali Sadri, president and chairman of the Wireless Gigabit Alliance. This consortium of tech companies has developed a specification known as WiGig (download PDF) that supports wireless communications at multi-gigabit speeds.
"With the emphasis on high-definition video, usage of the Web is changing," notes Sadri, who is also marketing director for Intel's wireless group. "Unfortunately, Wi-Fi hasn't kept up. Faster is always better."
Wi-Fi depends on the IEEE's 802.11 family of standards. 802.11a equipment, introduced in 1999, operates on the 5GHz radio band, while later 802.11b and g devices use the crowded 2.4GHz band. The most current dual-band 802.11n gear can operate on either band. Each 802.11 update has increased Wi-Fi's speed incrementally.
WiGig adds the 60GHz transmission band to the mix. With much more available spectrum than the 2.4GHz and 5GHz bands, the 60GHz band allows significantly faster throughput. (See Network World's informative video "How 60GHz will affect Wi-Fi" by Farpoint Group analyst Craig Mathias for details.) WiGig can operate at up to 7Gbps, more than an order of magnitude faster than today's 802.11n Wi-Fi, which can operate at up to 600Mbps.
In other words, WiGig delivers true multi-gigabit throughput -- enough to transmit an entire HD movie in a matter of seconds or smoothly stream it to a viewer. It also offers enough bandwidth to satisfy households with several data-hungry users connected at once -- such as a young child playing an online game in the den, a parent downloading a video-heavy work presentation in the kitchen and a teenager video-chatting with her boyfriend in the dining room.
WiGig could also be used to connect computers to peripherals, such as HD monitors or network hard drives, without a cable in sight. The new wireless spec is also compatible with current Wi-Fi devices.
Enhancing WiGig's speed is a cool trick called beam-forming. Unlike most wireless data systems, WiGig's signal doesn't spread out in a sphere, with most of it wasted. WiGig is smart enough to adjust the antenna parameters at both the sender and receiver to create a focused beam of data for a direct link that has minimal interference. Beam-forming technology is already being used in some Wi-Fi products, but unlike other Wi-Fi standards, WiGig actually relies on it.
"Beam-forming technology is very cool," says David Seiler, chief of the Semiconductor Electronics Division at the Commerce Department's National Institute of Standards and Technology (NIST). "It's what makes high-speed wireless like WiGig possible, and it will be used in a lot of other areas." Because WiGig is based on the same 802.11 specs as Wi-Fi, this technique can "extend the usefulness of Wi-Fi by five to 10 years," he adds.
There is a downside, though: WiGig's top speed has a range of only 45 feet. This will be a problem for home users who want to, say, connect a TV in the bedroom with a router in the basement, or for a business that wants to connect all of the employees in a small office wirelessly.
Sadri mentions two different ways to overcome WiGig's range limitation, neither of which is perfect. One possibility is to set up personal area networks (PAN) for each room or section of a home or office. That way, each PAN segment could pass along the data to the segment in the next room or section, although latency would increase each time the signal is relayed.
The other approach is a little more old-fashioned and involves installing gigabit Ethernet cables as a backbone for several WiGig transmitters placed in strategic locations throughout the building -- a solution that's likely more feasible for small businesses than it would be for home users because it requires running cables behind walls.
And, of course, WiGig will require a new generation of Wi-Fi routers and receivers that use the 60GHz transmission band. Armed with tri-band radios, these devices will also be able to operate on the 2.4GHz and 5GHz bands for interoperability with today's Wi-Fi equipment.
By the end of the year, Sadri expects four semiconductor companies, which he declined to name, to produce samples of WiGig's reference design chip for a new generation of wireless electronics. The needed chips should be in full production in 2012, he says.
By 2013, WiGig devices could be in TVs, computers, phones, tablets and other electronics, and eventually "a few uses we can't even imagine today," says Sadri.
Self-powered electronics: Power without the plug
Zhong Lin Wang dreams of electronic devices that can power themselves. If the materials science professor at the Georgia Institute of Technology's Nanoscience Research Group has his way, replacing or recharging batteries could soon seem "so 2010."
Wang's team at Georgia Tech has designed tiny generators that can produce enough energy to power very small devices. These high-output nanogenerators, HONGs for short, can produce between 2 and 10 volts from a flexible chip smaller than a fingernail.
The design starts with a microscopic array of zinc oxide (ZnO) fibers, or nanowires, each thinner than a human hair. These fiber arrays are embedded into multiple layers of metal electrodes and plastic polymers to create a flexible nanowire "sandwich."
Under an electron microscope, the strands look like the bristles of a very small brush, and they have the seemingly magical piezoelectric property of producing a tiny electrical current when moved or squeezed. Put billions of them together, and you get enough energy to power devices without using an external source of electricity.
"We turn motion into power," says Wang. So far, HONG devices have lit LED lights, run calculator LCD screens and powered rudimentary electronics in the lab. That's just the beginning.
Wang and his team are working on creating HONGs that can power complete wireless devices. Their current project is to make self-powered environmental sensors for a variety of uses.
For instance, the Georgia Tech team is working on a sensor that could be embedded in a bridge. "Surrounded by concrete, it wouldn't be easy to change the sensor's batteries," Wang quips. But with a HONG generator inside, the sensor could be powered by the vibrations of cars and trucks driving over the bridge.
The idea is that every 30 minutes, the sensor -- and dozens like it in the structure -- would send a reading to a receiving station for analysis. If the sensors showed that the bridge was in danger of collapsing, the structure could be shut down, preventing a disaster like the 2007 collapse of the Mississippi River Bridge in Minneapolis.
"This is an especially promising area," says NIST's Seiler. "It lets us think less about the device's battery running out of power and concentrate on what it's supposed to do."
While it's unlikely that nanogenerators will ever generate enough power to support large electronic devices like computers or TVs, a plethora of small devices could eventually be solely or partially powered by HONG chips. By 2013, Wang sees self-powered phones, digital music players and even a wireless keyboard powered by nothing more than the musician's keystrokes.
Wang says that the cost of adding a nanogenerator to devices would be low, because zinc oxide is a common material and HONGs are made with current semiconductor processing technology -- although some evolution of processing techniques will be needed. Additionally, HONG chips could lower the cost of certain products by partially or totally doing away with the need for a battery -- one of the most expensive components of any phone or music player.
Members of Wang's research group are also experimenting with more exotic, and potentially higher-power, piezoelectric compounds such as lead zirconate titanate, but they might be harder to process.
The bottom line is that for many of the electronic devices that surround us, the tyranny of the AC outlet and charger may end. "Motion is everywhere, waiting to be used for powering our future," says Wang. "All we need to do is harness it."
Wireless power: Electricity in the air
More than a century ago, electrical genius Nikola Tesla performed pioneering research and development on many of the things we take for granted today, from X-rays and alternating current electricity to efficient motors and generators and the precursors of radio. But when he turned his vivid imagination to sending electricity over the air with radio waves to power all sorts of devices and appliances without cords, it was an expensive failure.
Fast-forward to the present, where a company called Powercast is doing just what Tesla dreamed of: transmitting power via radio waves. "In a real sense, we're picking up where Tesla left off," says Harry Ostaffe, vice president for marketing and development at the Pittsburgh-based vendor. "We are sending power over the air for devices where it is expensive or inconvenient to change batteries."
Called power harvesting, the technique uses the company's book-size Powercaster transmitter to send either 1 or 3 watts of electricity into the air at the 915MHz radio frequency. At the receiving end, the power is pulled from the air by one of the company's Powerharvester chips, which convert RF energy to DC power.
At the moment, Powercast has two chips: One works best at close range and puts out up to 4.2 volts of continuous electricity for directly powering a very low-power device or charging a battery. The other can be used at longer distances from the transmitter to create an intermittent pulse of up to 5.25 volts for directly powering a low-power device.
These chips can grab small amounts of usable power -- from microwatts to low milliwatts -- out of thin air. That's not enough to run an MP3 player or phone, but it is sufficient for a device that uses very little power, like a Kindle e-book reader, according to Ostaffe.
Rather than developing systems for consumer electronics products, however, Powercast focuses on powering the various sensors that monitor our world, from temperature and pressure sensors in oil refineries to smoke alarms in homes and offices. The typical office today might have a door-position sensor that's part of its security system, a smoke detector, and a motion sensor to turn off the lights if nobody's in the room -- all of which could be powered wirelessly with Powercast's technology.
The company's goal is to eventually develop technology that can extract usable amounts of power from ambient sources such as Wi-Fi signals. But for now its system needs a transmitter to work, which means it's subject to range limitations, Ostaffe says.
"Our biggest enemy is the inverse square law," he jokes. This fundamental law of physics describes how energy radiating outward from a point source -- such as light or in this case radio waves -- is dissipated over distance. The energy available for the receiver falls off very quickly the farther you get from the transmitter.
"Right now, our usable range is about 40 to 45 feet," he says. That's long enough to cover part of a house or a few offices in a building. The typical office facility would probably need 10 or 15 transmitters per 40,000 square feet of floor space, set up around the building's periphery with antennas aimed toward the center.
NIST's Seiler agrees that RF-to-DC power has potential for certain kinds of devices. "It's promising but is limited by range and the amount of power an RF wave can hold. But it could power many smaller devices," he says.
Beam forming: The ability of a wireless communication device to adjust its antenna parameters and tailor its signal directly at the receiver for greater speed and reliability.
Electroencephalography (EEG): The monitoring of the electrical activity of the brain with several electrodes placed on the scalp.
Functional magnetic resonance imaging (fMRI): An offshoot of magnetic resonance imaging, this technology can pinpoint areas of the brain that are being used during an activity.
Inverse square law: A mathematical representation of how energy is dispersed over space, the inverse square law is useful in describing gravity, light, sound and all sorts of radiation. Energy falls off based on the square of the distance, so every time the distance is doubled, the energy available is one quarter of the original amount.
Piezoelectric materials: Piezoelectrics, like zinc oxide, are crystals that create electricity when squeezed or moved. They are used as sparkers in stoves, as sensors and as actuators.
Powercast isn't the only company developing RF-to-DC power technology. Nihon Dengyo Kosaku of Japan, for instance, has been working on a similar system that relies on a special rectifying antenna. Powercast, however, claims to have a head start on its competitors, saying it has chips ready to be integrated into products.
By 2012, Powercast hopes to have a household sensor product available to power smoke detectors that will never need to have their batteries changed. The company has also been working on shrinking its transmitters for home use. Ostaffe envisions a miniaturized transmitter "the size of a child's night light that would be powerful enough to power the smoke alarms in a modest-size home."
Nevertheless, distributing power over the air remains a tantalizing step or two away from mainstream availability. Although Powercast has the designs and chips ready, it needs a manufacturer to make and sell the actual products for consumers or businesses.
Self-healing batteries: Just-in-time repair
We all know the drill: You use your mobile device -- phone, tablet, laptop -- for a few years, then the battery dies and you have to replace it. Or you drop the device, the battery shorts out and you have to replace it even sooner.
And if a device isn't designed to allow the user to replace the battery himself (iPhones and iPads and other tablets are notorious for this), there's the extra hassle and expense of shipping the whole thing back to the manufacturer to swap out the battery.
But scientists at the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign have a better idea. Researchers led by professor Scott White are looking to extend the useful life of batteries in mobile devices, and they've figured out a way for the battery to fix itself, probably without the user ever knowing there was a problem.
Whether it's in a mobile phone or a notebook, a lithium-ion battery releases electricity by moving electrons from the lithium-based cathode to the cell's anode. The flow is reversed when you plug the device into a power source to charge the cells up for another cycle.
The problem is that over time, or if the battery is subject to a sudden shock, the battery's cells can be damaged, resulting in a power-killing short. At this point, all you can do is get a new battery.
White's team has tackled this problem by coating the battery's cathode with billions of microspheres filled with gooey gallium-indium. The key is that the spheres have been designed to break open when stressed (like when the device is dropped) or heated up (as when the battery shorts).
"We can trigger the microcapsules through mechanical force, temperature or pH," explains White from his lab at the university. "The capsules release their contents when damage occurs and a healing reaction takes place."
The gallium-indium quickly can flow in to fill in the gaps to fix the short, and the battery can be restored in as little as 40 microseconds. In most cases, that's not even enough time for the battery's control electronics to shut it down. "We get restoration of conductivity," adds White. "It is immediate."
For now, the microcapsule method works just once; if you drop your device a second time, you're out of luck. But White told me his team is working on ways to incorporate several different materials to make it possible to fix a battery several times.
The typical notebook battery gets about 100 to 150 charge cycles per year and lasts about three years before degrading to the point where it needs to be replaced. Using the Beckman Institute's techniques, five or six years is doable, and eventually we might see a 10-year computer battery.
"This is kind of cool, and very needed today," says Stanley Williams, senior researcher and head of the Quantum Science Research (QSR) group at HP Labs. "Batteries are the weak link in many of the products we use every day."
When might a self-healing battery be available? White, who was preparing a paper on the topic when we launched this story, is cautious, saying that the microcapsule method is still in the lab. Still, he says that adding microcapsules to a battery shouldn't interfere with the way they are made and shouldn't add too much to the cost of a cell.
Long term, he adds, the process could be used to fix all sorts of electronics, even an electric car's battery, effectively repairing it before we even know it's broken. Next on White's list are the power transformers and capacitors that make our electrical power grid work. If his group succeeds, someday we could conceivably see an end to power failures.
Neural computer control: Thoughtful computing
Despite losing to IBM's Watson computer on Jeopardy, the human brain remains the most powerful, flexible and complex information processor on Earth. But it has to interact with computers through our error-prone bodies. Click on the wrong icon or hit the wrong key and work can grind to a halt -- or worse, an afternoon of effort can be lost.
That's why scientists and other visionaries have long dreamed of interacting with computers through pure thought, using the brain to directly input, edit and manipulate ideas.
Like a scenario straight out of science fiction, using the brain as a computer interface is easily the weirdest and most speculative idea of the 10 breakthroughs we've covered in this two-part series. The reward is potentially huge, however. This capability could free us from the most inefficient part of the computing chain: the interface.
"It sounds crazy," says Dean Pomerleau, an Intel Labs researcher in Pittsburgh, "but you'd put on a cap that scans your brain and sit in front of your computer screen to check your calendar, reply to annoying emails and work on that big spreadsheet from work -- all without typing or moving a mouse."
Starting at the University of California, Los Angeles, in the 1970s, a long line of researchers around the world has experimented with brain-computer interfaces (BCI), first using animals and later humans as well. Many of these efforts have involved implanting electrodes inside the brain or on its surface. One problem with that approach is that scar tissue tends to develop around such implants and it interferes with the signal. Other projects have attached electrodes to the subject's scalp, but the skull can block or distort the brain's signals.
Despite these limitations, scientists continue to move BCIs forward. Pomerleau, for example, is working with researchers from the University of Pittsburgh and Carnegie Mellon University on a project known as NeuroSys.
For this group, efforts to turn thought into computing action began with observations of people's brains using a functional magnetic resonance imaging (fMRI) machine. Subjects were told to think about specific words like "search" or "dog," and the machine created an image of the neural activity, lighting up the areas of the brain that were creating the thought.
Working with many test subjects (English speakers only for now), the NeuroSys researchers started with nouns and moved on to verbs, amassing brain scans and noting similarities among them until clear patterns emerged. All this data has been incorporated into a computer program that can translate neural activity patterns to words.
The program has built up a vocabulary of 1,000 words and can parse simple sentences from subjects' brain patterns. Of course, the project needs to deal with the frustrating nuances and ambiguities of language, but it's been surprisingly successful. So far, the program is 90% accurate in predicting what subjects are thinking, according to Pomerleau.
The problem is that there are few computer users who have the desire, or the financial wherewithal, to sit in a $2 million brain scanner to compose a memo to the boss about a new marketing campaign. "A big leap is needed in the sensing technology, to a point where it can be miniaturized," says Pomerleau.
That leap may be at hand. Small electroencephalograph (EEG) sensors that track and interpret brain activity can be built into a headset or cap and may prove be a good stand-in for interpreting fMRI readings.
Primarily used in medical research, such devices are also appearing in everything from "neuromarketing" aids (wireless headsets that register test subjects' responses to marketing and branding) to crude toys that, for instance, let you pretend you're a Jedi knight by controlling a ball's height with mental power. Recently, a group of German engineers operated a specially modified car with one.
[ Related reading with cool videos: A mind-blowing look at today's mind-controlled, mind-reading technologies ]
On the computer interface front, Austria-based G.tec showed its Intendix system at this year's CeBit show in Germany. It uses an EEG cap studded with electrodes in conjunction with software you load on a Windows PC. The interface is laid out like a typical qwerty keyboard, with a few additional symbols for things like printing and sending email. After training the system, all you do is stare at the Intendix screen and think about the letters, numbers or symbols to spell out your message.
At the moment, the system can recognize about five characters a minute -- not exactly speed typing, but it's a start. Because of the system's $12,000 price tag, brain-powered computing will probably first be used for people with limited voluntary muscular control or "locked-in syndrome" diseases including amyotrophic lateral sclerosis (ALS), a.k.a. Lou Gehrig's disease.
A neural interface would open a new world for them, and eventually for the rest of us. But it could be decades before the technologies become advanced enough -- and inexpensive enough -- to make sophisticated brain-computer interfaces mainstream.
"The payoff here could be huge," says Supratik Guha, director of the physical sciences department at IBM. "But there's a lot of work that needs to be done to make this type of interface work."
Read Part 1 of this series: 5 tech breakthroughs: Chip-level advances that may change computing