3 October 2014

Future Military Sensors Could Be Tiny Specks of ‘Smart Dust’

New technologies allow for extremely small—and ubiquitous—military sensors

In the 1972 science fiction story The Unknown by Christopher Anvil, three space pilots find themselves plagued by “ultra-miniature spy-circuits.” Tiny computers used for espionage and no bigger than a speck of dust.

“They drift in like dust motes,” one space pilot says. “But you have no control over where they drift. An air current, or a static charge, can completely foul up your arrangements.”

In 1972, dust-sized electronic spies were far-out stuff. But in 2014, it’s not so far out at all.

Coined by University of California, Berkeley professor Kristofer Pister, the term “smart dust” refers to tiny electronic bundles of power, sensors, computing and communications electronics that are cheap and abundant enough to scatter like, well, dust.
These tiny machines sense their environment, perform basic data processing and communicate with each other—to serve medical, industrial and military purposes.

Once a science-fiction idea, the smart dust concept caught the imaginations of geeks and investors a few years ago. When Pister coined the term in the late 1990s, the concept attracted funding from DARPA, the Pentagon’s advanced research outfit.

It’s easy to see why the military wants smart dust. During the Vietnam War, the military deployed a slew of remote sensors to detect and track North Vietnamese troops. Although the planned “McNamara Line” along the DMZ never fully materialized, the sensors proved their worth during the Siege of Khe Sanh in early 1968.

Better sensors airdropped over the Ho Chi Minh Trail—capable of detecting motion, sound, metal and even smell—guided the massive U.S. air campaign against North Vietnamese supply lines.

But these sensors were huge in comparison to what’s possible now. Over the past dozen years, miniature sensor network technology progressed enough to bring specialized commercial devices to market. Used to monitor high-value packages, monitor machinery and building environments, these units are more cigarette-box sized than dust-sized.

A graphene structure like this could be a smart dust antenna. CORE-Materials/Flickr illustration. Top—Graphene structure.
University College London/Flickr illustration

Teeny tiny sensors


In 2001, Pister and his colleagues conducted a field demonstration for DARPA at the Marine Corps’ base at Twentynine Palms. A small drone dropped six “motes” the size of a pill bottle near a road.

After synchronizing with each other, they detected the presence, course and speed of a Humvee and a heavy transport truck. When the drone passed overhead, the motes transmitted their data to the drone, which then beamed the information down to a base station.

Last year, a University of Michigan team showed off its Michigan Micro Mote, a solar-powered wireless computer not much bigger than a coarse grain of salt. A few years ago, Hitachi showed off experimental radio-frequency chips the size of dandruff flakes.

There are still big engineering constraints on such tiny devices—such as how to provide enough power and how to broadcast communications signals. But new approaches may solve these problems.

Antennas need to be of sufficient size enough to operate, but it’stheoretically possible to build antennas from graphene only a few atoms thick. Solar cells also require enough space to collect energy, and sunlight isn’t reliably available.

However, tiny motes require very little power.
Clever software algorithms can also dramatically extend operating lives. This can mean turning on a mote for milliseconds, transmitting its sensor data before shutting off. This saves lots of energy over time.

Tiny vibration motors can harvest power from anything that moves or shakes—strides, heartbeats and active machinery. Flaps, weights and supports created using chip-fabricating techniques turn mechanical vibrations into power for as long as the machine holds up. Perhaps years.

Even nuclear power can play a role in energizing smart dust. During the 1970s, plutonium powered a lot of pacemakers sewn into human chests. A tiny, long-lived power source meant fewer dangerous surgeries to replace the pacemaker’s juice over a patient’s lifetime.

Though huge by nano standards, these tiny medical devices are still tickingdecades after their manufacture. But better chemical batteries and worries about tracking thousands of little nukes—not health concerns—led to the withdrawal of nuclear-powered pacemakers from the market.

In 2005, University of Wisconsin researcher Jake Blanchard demonstrated a tiny system that used the electron-spitting quality of radioactive nickel to attract a springy, hair-sized lever of copper. When the copper lever shorted the circuit, it sprang back and generated an infinitesimal amount of electricity.

Scan Eagle drone on amphibious dock landing ship USS Comstock on Feb. 26, 2011 Navy photo

Dusting off

Smart dust requires cheap, plentiful motes able to withstand environmental challenges. However ingenious their design, they must be cheap as dust to or their cost will be prohibitive. But long-term trends in microelectronics suggest these low costs are on their way.

Dust is also dirty, irritating and sometimes toxic. Smart dust can’t be ubiquitous like real dust—we’ve already got enough of that stuff to worry about. That’s why DARPA is thinking about how to get smart dust to die offon its own. This means intentionally restricting its battery life.

Challenges to the baseline technologies—power, communication and sensing—is harder work. There’s also the possibility an enemy electronic warfare team could scramble the tiny sensors.
A widespread sensor mesh of smart dust that relies on radio waves to communicate will be just as vulnerable to electronic countermeasures as any other radio-based technology.

But smart dust might overcome such obstacles by using swarm behavior.

Thousands of motes, each generating a tiny bit of laser light, could transmit information to overhead drones by working together as a big distributed “flashlight.” Or they could link up into a big low-frequency radio antenna to beam out their signals.

Ultimately, it might be the software behind smart dust that becomes the smartest thing about it.

Which makes sense. The security, networking and processing algorithms able to bind together the motes in our minds’ eyes are the key to this promising technology, just as our smartphones are nothing but shiny bricks without the software and cell networks behind them.

But the worst thing that could happen to your smartphone is dropping it on a sidewalk. Smart dust doesn’t have this problem. Though as Anvil hypothesized in his science-fiction story, ultra-miniature spy sensors might just drift away in the wind instead.

No comments: