| Embedded
Collaborative Computing
Smart sensing systems
are becoming increasingly
available for a
variety of commercial
and national security
applications. PARC researchers are addressing
fundamental problems
involved in designing,
programming, and
deploying distributed
sensing and diagnostic
systems, both in
the environment
and inside machines.
Work in this
area is building
on
rapid advances
in several converging
technologies, including
MEMS, wireless
networking, and
embedded processing.
These advances
have enabled the
deployment of large
numbers of inexpensive
micro-scale sensors.
Measuring changes
in heat, sound,
vibrations, temperature,
humidity, electrical
current, and other
physical phenomena,
these sensors can
collect more information
than has previously
been possible.
Systemic Approach
ECC’s
systemic approach
leverages
this information-gathering
capability and
the connectivity
between distributed
sensors. This unique
approach distributes
computation over
a large number
of processors in
wired and wireless
networks, and employs
model-based reasoning
techniques to analyze
the information.
A
multidisciplinary
team of scientists
is bringing expertise
from a variety
of disparate areas – including
signal processing,
software engineering,
computer science,
artificial intelligence,
distributed algorithms,
hardware prototyping,
large-scale experimentation,
and MEMS devices – to
bear on ECC research.
The team is developing
tools, algorithms,
prototypes, and
experimental systems
that demonstrate
feasibility and
proof of concept
for new generations
of smart sensor
networks.
ECC
research is focused
on two
problem spaces – collaborative
sensing, which
deals with networks
distributed across
large geographical
distances, and
distributed diagnostics,
which focuses on
networks of sensors
located inside
machines.
Collaborative
Sensing
Algorithms
developed
for battlefield
scenarios
can be reused
for monitoring
of wildlife,
environmental
pollutants
and power
grids. |
Large-scale,
distributed,
sensor-rich wireless
networks are
designed to track
physical
phenomenon, including
multiple moving
objects such
as vehicles or
animals.
Potential
applications
include traffic
control, battlefield
target tracking,
security, and
monitoring of
wildlife,
environmental
pollutants,
and infrastructures
such as
power and telecom
grids
Scientists
are exploring the
problems of information
processing, communication,
storage, and routing
in such networks,
which are constrained
by energy and bandwidth
limitations.
Distributed Diagnostics
Sensor-rich
networks that track
the
performance of
components inside
electro-mechanical
machines are leading
to a new generation
of machines that
can diagnose and
repair themselves.
Distributed sensors
monitor multi-modal
data, measuring
such physical phenomena
as vibration, noise,
electrical current,
and its signature – changes
in its signal over
time. Researchers
are developing
scalable, model-based
techniques for
processing the
information from
these distributed
sensors to achieve
highly distributed
sense making, diagnosis,
and rapid device
reconfiguration
and repair.
|
