sensors and sensor data fusion are driving new Smart System innovations
enabling a whole new generation of applications that are self-sensing,
self-controlling, self-optimizing and “self-aware”—automatically,
without human intervention. As networks continue to invade the
“physical” world, solution designers are seeing the new values that
come from the growing interactions between sensors, machines, systems
SENSING TINY THINGS
you understand how fast the world is changing? One indication of the
speed is the fact that we’re drowning in unprocessed information. We’re
creating data at 2X the rate we’re deploying traditional bandwidth to
carry it, but almost all the data created to date has never been
analyzed. More than half the data created by physical or operational
systems loses any value that could be derived through analysis in less
than a single second.
Yet according to the National Science Foundation, there will soon be trillions of
sensors on the earth. And forecasters predict that in just a few more
years there will be more processing power in smart phones than in all
the servers and storage devices in data centers on the earth today.
Ready or not, we’re rushing into the future of truly distributed
systems and intelligence.
always begin by sensing small things and drawing inferences. For
example, there is great value in knowing how people use “white goods”
like home appliances. If you embedded a microprocessor in the plastic
of an electrical outlet, you’d have true local processing as opposed to
processing in a remote cloud. From there, you could infer almost
anything by sensing the electrical current “signature” and its usage
profile—not just energy used, and whether a washing machine’s motor is
about to fail, but the fact that the consumer just washed a load of
whites versus a load of colored clothes. If you had an inkjet printer
plugged into that same outlet, you could know whether the consumer was
printing colored pages versus black-and-white.
of data from a “sweat patch” for measuring human perspiration can
reveal the emotional state of an athlete wearing it, as well as the
level of physical stress she’s under. If a worker is wearing the patch,
you can see if they’re being exposed to the many dangerous substances
that exist in factories, farms, and other workplaces.
is another complex phenomenon that can be greatly understood by
correlating data values collected from simple pressure, temperature,
and moisture sensors with additional spatial and temporal parameters
that place the data into a richer context. I could be running a fleet
based on weather forecasts, but if I add the data from sensor packs
mounted directly on my vehicles, I’m less likely to be impacted by
unexpected weather conditions. The closer that systems are to
real-time, the more efficient and cost effective they can be. All such
data, with its related context, has extraordinary value to everyone.