NTT DoCoMo in Japan, one the world's leading mobile providers, recently announced a prototype wireless network that could send data packets at 2.5 gigabits per second -- fast enough to download a DVD movie in between 7.5 and 10 seconds -- to a mobile device traveling at 20 kilometers per hour.

If their prototype wireless technology can produce even a fraction of that 2.5-gigabit transfer rate in real-world applications, it would vastly enhance mobile functions -- allowing video telephony, robust Internet connectivity, and streaming media services, while at the same time extending the range of traditional voice calls.

These high-speed data networks, along with increasingly powerful mobile handsets, have the potential to supplant the use of desktop computers -- a trend that's already occurring in some Asian countries. This potential market has DoCoMo, along with almost every other major wireless player, including Motorola, Samsung, and Qualcomm, scrambling to develop their own technology for the next generation of wireless networks, often labeled "4G."

DoCoMo's demonstration gives a glimpse into the two types of technology that will most likely be adopted to increase bandwidth and range: MIMO, which is applied to network base stations and mobile devices, and QAM, which loads more data onto radio waves.

MIMO (multiple input, multiple output) uses multiple antennas to send and receive data, as well as specific coding that scrambles and unscrambles the signals produced by those antennas (see "Faster, Farther Wi-Fi"). A base station that uses MIMO technology has multiple antennas that simultaneously receive and send data to and from wireless devices. Unlike base stations with a single antenna, those with MIMO use the multiple antennas to create a number of intertwining channels through which data moves. The jumbled signals are untangled by a "signal processing" that sorts through the bits.

Because MIMO base stations can handle many more data streams than single antenna wireless stations, there's more bandwidth and built-in redundancy, which increases network reliability and range, says Rob Gilmore, senior vice president of engineering at NextWave Wireless. By deploying MIMO routers, a mobile network such as DoCoMo's 4G can increase the amount of data sent and received, as well as increasing the range, he says.

Most MIMO routers have two or three antennas. In DoCoMo's demonstration, the router as well as the receiver used six antennas to produce rates of 2.5 gigabits per second, says Satoru Kawamura, a company representative. Tripling the number of antennas on a MIMO access point and receiver can triple the amount of bandwidth of the network, says Gilmore.

Researchers at Harvard University have shown that nanowire transistors can be at least four times speedier than conventional silicon devices. The principal researcher, chemistry professor Charles Lieber, says this could lead to inexpensive, high-performance, flexible electronic circuitry for cell phones and displays. It could also save space and further increase speed, he says, by allowing memory, logic, and sensing layers to be assembled on the same chip.

Nanowires have been considered a promising contender for use on future logic chips because of their very small size (about 10 nanometers wide) and because they can be made without complicated lithography, says Peidong Yang, professor of chemistry at University California, Berkeley. Until now, though, the performance of nanowire-based transistors has lagged far behind that of other potential nano devices, such as carbon nanotubes, and even conventional devices. But the new Harvard research suggests that nanowires have surpassed conventional transistors and nearly caught up with nanotubes.

This may give nanowires an edge over carbon nanotubes (see "Carbon Nanotube Computers"). Nanowires are made with regular crystal structures and uniform electronic properties -- a level of predictability essential for manufacturing high-performance electronics. Nanotubes, however, come in batches of different sizes and structures, each of which can perform very differently -- so until a good sorting method can be found, it will be difficult to use nanotubes in high-end processors.

The first applications for nanowires will likely be ultra-sensitive sensors for single molecule detection (when molecules bind to the nanowires they create a detectable change in the current flowing through the wires). Such applications could be ready in two to three years, Lieber says (see "Drugstore Cancer Tests").

Nanowire transistors may never replace more conventional devices in computer chips used in laptops and personal computers -- the cost of developing large-scale manufacturing would probably not be justified by a 4 to 5 times improvement in performance, Lieber says. But, he adds, the new performance figures suggest it will be well worth scaling up the technology to manufacture them for applications where the ability to assemble nanowire transistors at room temperature on various surfaces, including plastic, will bring an added advantage. For instance, in flexible displays nanowire transistors could be used to embed information-processing in the screen itself.

The technology might also be useful for extremely compact devices, since it would be possible to layer memory, logic, and even sensing circuitry on top of each other, rather than side by side or on separate chips. The nanowires are applied to chips and connected to the source, drain, and gate using room-temperature processes, allowing consecutive layers to be applied without damaging previous layers. "If you can put ultra-high-performance materials into 3-D structures, through layer by layer assembly, it allows you to put a lot more stuff into an area," says Lieber. The proximity of the layers, a mere 100 nanometers apart, could also speed performance, he says.

One of the qualities that distinguishes this current work from earlier nanoscale electronics research, including his own, Lieber says, is that the measurements used are industry standards, which makes it possible to compare how nanowires would perform in real devices.

The key to the improved performance is a "core-shell" structure of the nanowires, which confines electrons, or their counterparts, electron holes, in a small space. That allows electrons to zip through the wires quickly, which is key to the speed improvements. In a recent paper in the journal Nature, Lieber made nanowires with a germanium center surrounded by a thin coating of crystalline silicon. And in work described in Nano Letters, the researchers showed the versatility of nanowires by using gallium nitride, which could be useful for high-power, high-temperature applications.

"These two papers come up with very interesting ideas for using this core-shell structure to enhance the performance of these transistors and basically make them much more robust and reliable," says Berkeley's Yang. With nanowires, he says, "you get very small features and different compositions, and you also have access to all these nonconventional heterostructures, like these core-shell structures, that enable you to engineer the electronic structure. These are not things you can do easily with conventional technology."

A reconfigurable chip developed by ChaoLogix in Gainesville, FL, makes it possible to morph a circuit from one type into another in an instant. Having the ability to effectively redesign chips an unlimited number of times after they've been manufactured could make chips faster and more robust. And, ultimately, it could bring down the cost of producing integrated circuits, by reducing the need to make expensive, custom-built chips.

The novel chips work by exploiting inherent "chaotic" behavior within the integrated circuits, enabling a single, simple circuit to behave like any kind of logic gate. Such a chip could be transformed, for example, from a graphics card into a memory chip and back again -- in just two computer clock cycles. "We have blurred the line between software and hardware," says William Ditto, chief technology officer of ChaoLogix, which was spun out of research at the University of Florida.

In many respects, the concept is similar to the development of software-defined radios [SDRs], says Ditto. These are devices that use a combination of custom-built integrated circuits and existing reconfigurable chips to provide a flexible mix of hardware and software, to make wireless devices that can adapt to operating at different radio frequencies and standards. But whereas SDRs can make only radio devices and consist of several chips designed to perform wireless functions, ChaoLogix's chips could, in theory, replace all of these chips in a single device.

Existing reconfigurable chips, called field programmable gate arrays (FPGAs), contain programmable interconnects that can be rewired to perform different functions. But FPGAs are relatively slow to reconfigure, typically taking milliseconds for each rewiring, or about one million times slower than ChaoLogix's chips.

Because of this limitation, FPGAs tend to be reconfigured only once to form a single permanent circuit, usually as relatively cheap alternatives to building dedicated chips. "Making a dedicated chip is very expensive," says Allan Cantle, CEO of Nallatech, in Glasgow, Scotland, which develops software for FPGAs. "You can easily spend tens of millions of dollars just making your first working chip."

Rather than using programmable interconnects, ChaoLogix's approach is to use fixed circuits and instead exploit their inherent "noise" or chaos to make them produce different outputs without changing them. Normally, the circuits on a chip consist of arrangements of transistors designed to behave like a specific type of digital logic gate, such as a NAND and NOR gate. But if the inputs voltages to these circuits fall below certain thresholds, their behaviors become chaotic, producing undesirable outputs.

ChaoLogix's trick is to put these chaotic states to use. They've designed a logic gate circuit that's capable of behaving like any kind of logic gate -- if the input voltages are just right.

The common notion that chaotic systems are unstable and unpredictable is not accurate, says Ditto. Such systems can be extremely sensitive to changes, and it is possible to produce desired states reliably and reproducibly provided you ensure only minor changes are made to the inputs.

"Just making small changes to the input, you can adapt [a circuit] to do totally different things," says Celso Grebogi, professor of nonlinear and chaotic systems at University of Aberdeen in Scotland. This creates a greater degree of flexibility, because it makes more states available in a given system, he says. Because of this, Grebogi sees engineers increasingly turning toward chaos to get more out of their hardware.

Standing bits on end, for example, requires the use of a soft underlayer; this layer, which guides the magnetic flux from the drive's read-write head as bits pass below it, makes the head more efficient, which means it can use a stronger magnetic field, which, in turn, means the grains can be made from a magnetically "stronger" material that's less vulnerable to the superparamagnetic effect.

Perpendicular magnetic recording has been a long time coming -- in part, because it involves arranging new materials in tricky new configurations, and in part because manufacturers wanted to test the technology thoroughly before putting it in front of consumers. "When you introduce perpendicular recording there is a fairly significant change in the material structure of the disk," explains Best. "There's a soft underlayer which carries the current from the read/write head, and a media layer on top, and when you introduce these materials you have to worry about things like surface roughness and susceptibility to corrosion from humidity. We built over 20,000 drives and put 5,000 of them into extended reliability tests before we even shipped the product. We wanted to make sure it was super-reliable."

"Perpendicular recording is not an easy technology to bring to market," confirms Maciek Brzeski, vice president of marketing for Toshiba's Storage Device Division. "All of the prior technological improvements on hard drives have been somewhat incremental. But with perpendicular recording, it's not a slam dunk. It takes some time and persistence."

Manufacturers estimate that perpendicular recording will allow them to keep shrinking bits until hard drives reach 500 gigabytes, or perhaps 1,000 gigabytes (one terabyte). Eventually, however, the superparamagnetic effect will come back into play. "It's just bought us a little bit of time -- a few generations of hard drive models -- and then we get into the same problem," says Best. He says Hitachi is already studying "patterned media" --the idea of arranging individual grains in specific patterns to reinforce their magnetic stability--as a way to break the 1-terabyte barrier.

Toshiba will begin mass production of its 2.5-inch drive in August, and expects to see them in high-end consumer laptops in 2007. Hitachi says its 160-gigabyte, 2.5-inch drive will show up in a broader range of laptops; it plans to release a 1.8-inch drive for handheld devices next year.

Auspiciously for manufacturers, the new perpendicular-recording drives are arriving at just the moment when digital video -- which consumes large amounts of disk space -- is booming. Consumers can now use their computers to download network TV shows from iTunes, YouTube, Google Video, and other sources; to record directly from cable or satellite connections, as TiVo and other digital video recorders do; and to upload video they've captured themselves on the newest digital video cams, which can eat up 15 gigabytes of space per hour of video.

"For years we've dreamed about the consumer applications starting to provide the growth engine for hard drives, and now it's actually here," says Hitachi's Best. "So it's a pretty exciting time."

Cellular service providers are deploying an advanced version of the high-speed, mobile 3G data services already offered on many of their networks in an attempt to accelerate the adoption of broadband-enabled wireless computing. By doing so, they're hoping to fight off a growing challenge from alternative Wi-Max networks, which deliver data (and increasingly voice calls) over long distances -- potentially threatening the core business model of these companies.

Already, 18 cellular networks worldwide have launched or upgraded a version of the High-Speed Downlink Packet Access (HSDPA) network in 14 countries, with another 29 nations expected to be upgraded by the end of this year, according to the Global Mobile Suppliers Association. The networks promise respectable speeds of 1.4-3.6 Mbps -- the limit of today's mobile phones and PC cards -- although the HSDPA technology can actually support up to 14 Mbps.

Philippe Keryer, president for mobile radio activities for Paris-based network equipment vendor Alcatel, says that speed is critical because mobile providers don't want to leave customers disappointed (as many were with 3G networks). "People are inevitably going to compare this to their DSL line," says Keryer, so operators "need to have service that can compare."

While it's taken the cellular companies some time to commit to HSDPA networks -- equipment makers have been eager to sell them on it since 2004 -- the changeover, once initiated, is a quick process. That's because HSDPA is primarily a software upgrade on the network end.

Bill Krenik, manager of wireless advanced architectures for Texas Instruments, which is a major wireless semiconductor supplier, says the fundamental difference between HSDPA and current 3G technology is that the equipment operating the antennas that communicate with the mobile devices adapts on the fly. Rather than broadcasting one form of signal, they tailor their signals to each user based on the quality of the link. "The base station is going to assess the quality of the channel and use the coding and signal that's most suitable," says Krenik.

When the connection is good, for example, an HSDPA base station will transmit a more efficient signal, called 16-QAM, where each "symbol" transmitted and received corresponds to 16 bits of information -- four times more per symbol than a 3G transmission. Over a strong connection, the base station will also transmit less error-correcting information, the coding that enables the phone to check whether it got a clean feed; a tighter code means faster transmission of the data packets the user wants.

And when a phone does detect a problem and asks the base station to resend garbled or lost data, HSDPA responds much faster than 3G. In 3G networks, such "retransmission requests" initiate a long, slow hunt for packets at various layers in the network. In HSDPA, the outgoing data is buffered in memory added to the base stations, which can thus respond immediately to the retransmission requests. Krenik says this feature, by reducing network traffic, can double data throughput on a cellular network.

In the hype-filled world of nanotechnology, Phaedon Avouris, head of IBM Research's nanoscience and technology group, has a reputation as a meticulous and somewhat skeptical scientist. By his own description, he is one of those researchers whom reporters call to get a "realistic assessment" of the latest nanotech breakthrough. These days, though, the IBM chemist sounds uncharacteristically upbeat.

The reason for his excitement can be seen in a microscopic image recently produced in his lab. It shows a thin thread draped over several thick gold electrodes. What is not so apparent is that the thread, a single carbon nanotube, has been modified and positioned so that it forms two types of transistors, each a few nanometers (billionths of a meter) in diameter- a hundred times smaller than the transistors now found on computer chips. What's more, the nanotube transistors work together as a logic gate, the fundamental computer component responsible for selectively routing electrical signals, transforming them into meaningful ones and zeroes.

The IBM device is one of the first examples of electronic circuitry constructed out of individual molecules. And while it's merely a crude laboratory demonstration, its successful fabrication is nevertheless a further tantalizing clue that carbon nanotubes could one day replace silicon crystals as the building blocks for ultrafast, ultrasmall computers. More measurements are needed, says Avouris, "but our current results show, after taking into account difference in size, nanotube transistors show a performance superior to that of state-of-the-art silicon transistors."

Indeed, carbon nanotubes are, in theory at least, the ideal material for building tomorrow's nanoelectronics. And now, a little more than 10 years after their discovery, nanotubes seem ready to make the transition from exotic laboratory wonders to materials useful in actual technologies. Prototypes of nanotube devices are being tested in everything from full-color flat-panel TV screens to ultrabright outdoor lighting to a simpler, smaller x-ray machine; consumers could be shopping for a flat-screen TV that uses nanotubes as early as Christmas 2003.

But it is in computer memory and logic that nanotubes could have their greatest impact. Microelectronics now use silicon transistors with features as small as 130 nanometers across, which means that Intel can squeeze some 42 million of these transistors onto its Pentium 4 chip. However, it's getting harder-and far more expensive-to continue to shrink silicon devices. Using nanotubes or related materials called nanowires as tiny electronic switches would allow computer designers to cram billions of devices onto a chip. If these molecular transistors work-and that is still a big if-replacing silicon will likely take years. But the ambition, says Charles Lieber, a Harvard University chemist, is to build electronics with performance "orders of magnitude beyond silicon. We're trying to break with what is being done, to really change things."

While developing this method of organizing nanotube transistors is an important step, much work remains to be done before commercial processors will be available. For one thing, exploiting the full potential of nanotube transistors will require improving the leads, possibly by using nanotubes in place of the palladium wires.

But perhaps a more pressing problem is finding reliable and inexpensive ways to isolate different types of carbon nanotubes. Current fabrication techniques produce a mix of nanotubes with different sizes and electronic properties, not all of which will work well in integrated circuits.

Because of these challenges, the first applications of carbon nanotube transistors will probably not be as high-performance processors, Hannon says, but highly sensitive sensors that work even with a mix of different nanotubes.

Meanwhile, others are developing devices that don't rely on nanotubes' high-end electrical properties, but rather on features such as their strength and flexibility. This skirts the need both to sort and to individually arrange the nanotubes. The Woburn, MA-based company Nantero, for example, takes advantage of nanotubes' strength and flexibility to make memory devices. "We use [nanotubes] as electromechanical devices, so we just bend them up and down to represent zeros and ones," says Nantero CEO Greg Schmergel. In this application, clusters of nanotubes rather than single tubes can be used, so they can be patterned using lithography.

Eventually, Schmergel says, nanotubes could replace every part of semiconductor devices by using all of the tubes' features. "Nanotubes have quite a number of unique properties all combined in one material. They can replace memory, logic, the interconnect, ultimately they can replace everything in the chip, so it definitely makes sense to pursue all of those angles," he says.

Amid increasing competition from Advanced Micro Devices (AMD), Intel is changing its chip-making philosophy: it's paying more attention to the power requirements of its microprocessors.

In July 2006, the chip-making giant will release a new microprocessor, called Core 2 Duo, designed for laptops and desktops. The new chip is based on Intel's current chip architecture, which replaced traditional single-core processing with two processing centers on a single chip. The company says that the Core 2 Duo will perform better than its current dual-core chip, and will be more energy-efficient, which could make laptop batteries last longer and desktop towers run cooler.

Paying attention to power consumption in microprocessors is a relatively new concept for the company, says Steve Pawlowski, a senior fellow at Intel, adding that the move may help Intel regain market share from its rival AMD. Historically, the most important metric in the industry has been processor performance -- the speed at which a processor can complete a task, such as calculating a spreadsheet. "We've always focused on performance at the expense of power [use]," Pawlowski says.

But basic changes have occurred in the PC market, which first led AMD, and now Intel, to rethink microprocessor designs. First, mobile devices have become the primary PC for many consumers -- who don't want a device that quickly drains a battery or gets too hot. Furthermore, as the size of transistors shrink, they're more likely to waste electricity through a physical process called "leakage," says Kevin McGrath, an AMD fellow -- and the more transistors on a chip, the more electricity is wasted.

AMD has been working on more-efficient microprocessors for several years, and now Intel is trying to level the playing field. Both Intel and AMD have tackled part of the problem by converting their chip line-ups to dual-core processors (see "Multicore Mania," December 2005), which turns out to be one way to increase efficiency. "Interestingly, going to multiple cores can be a very power-efficient way of computation," says Milo Martin, professor in the computer and information sciences department at the University of Pennsylvania.

Three aspects of multicore chips make them more efficient. First, when a chip has more than one core, the speed at which each core computes can be slowed down without impeding the speed of the entire chip. By slowing down the clock speed, explains Martin, engineers can decrease the computational rate of a single core by a factor of five, from one gigahertz to 200 megahertz, and the core consumes only one-30th of the power. Then, he says, even if five of those cores are assembled onto a single chip, only one-sixth of the power is consumed, yet the total computational rate of one gigahertz is maintained.

TR: Skype, the most popular VoIP client, already has encryption software, so why doesn't Zfone work with Skype?

PZ: Skype is not compliant with VoIP standards; they have a closed protocol. Skype uses its own encryption software and it doesn't tell anyone how it works. I prefer to use encryption that is open; I publish my source code.

TR: Who would use this VoIP encryption software?

PZ: Who wouldn't use this? Who wants to not be wiretapped? I'm not talking about wiretap from law enforcement -- I'm talking about wiretap from organized crime. Organized crime is doing phishing attacks and taking over your computer with hostile software. I'm making the prediction that those same criminals will attack VoIP when it gets big enough. It could be point-and-click wiretapping from the other side of the world.

TR: With your e-mail encryption software, you were under a criminal investigation by the U.S. government, which alleged that you violated export restrictions for cryptographic software. The case was eventually dropped, but how do you plan to avoid such a complication this time around?

PZ: This time I'm being careful about getting good legal advice and following export controls. I've filled out the paperwork and filed with the U.S. Commerce Department. I'm getting things back that clear it for export. I'm being very careful this time.

TR: Your software release is timely in light of the ongoing news about NSA's program to collect information about phone calls in the United States. Could you discuss the tension between technology and the law, especially when it comes to emerging forms of communication and keeping information safe and private?

PZ: I don't see this as a black-and-white situation. I sympathize with the need for NSA to catch the bad guys, and I want them to catch these bad guys. But we have to be careful about creating surveillance machinery that may be used for other purposes later.