RFID Sensor Networks
Joshua R. Smith
Associate Professor
Computer Science & Engineering and
Electrical Engineering
University of Washington
FISC 2030
March 22, 2012
RFID Sensor Networks
Joshua R. Smith
Associate Professor
Computer Science & Engineering and
Electrical Engineering
University of Washington
FISC 2030
March 22, 2012
WISP 3 axis x 10 bit accelerometer
WISP: Wireless Identification
and Sensing Platform
First UHF RFID w/
accelerometer
RFID Sensor Networks with the Intel WISP Winner Best Demo, Sensys 08
M. Buettner, B. Greenstein, R. Prasad, A. Sample, J.R. Smith, D. Yeager, D. Wetherall.
WISP: A Passively Powered UHF RFID Tag with Sensing and Computation, D.J. Yeager, A.P.
Sample, J.R. Smith, in S.A. Ahson, M. Ilyas Eds. RFID Handbook: Applications, Technology,
Security, and Privacy, CRC Press.
WISP & UHF RFID
UHF RFID: Much longer range than earlier generations of RFID (up to 30 feet)
Uplink communication by backscatter (reflection)very low power
Backscatter energy efficiency improves with Moores law, unlike conventional radio!!
WISP has no batterypowered by reader device
WISP: Programmable, software defined, sensor-extensible RFID tag
Tag
RFID reader
ant
Power & data (downlink)
Data (uplink)
Tag
WARP Ambient RF Harvesting
WARP: Wireless Ambient Radio Power
Harvests from environmental RF sources
Television band
GSM band
Onboard sensing and computation
Rectified
voltage
Light level
Low-power radio uplink
8pm
9/8/10
8am
8pm
9/9/10
8am
4pm
Wireless Ambient Radio Power, A.P. Sample, A.N. Parks, S. Southwood, J.R. Smith in
Wirelessly Powered Sensor Networks and Computational RFID, Springer, 2012
1 MW TV transmitter 4.2 km distance
WARP Ambient RF Harvesting
Battery-free environmental sensing
Weather statistics
Toxic gas monitoring (e.g., CO)
Next Steps
Extended range radio
Optimized harvesting and storage
GSM Harvesting
Printed Low Power Amperometric Gas Sensors Employing RF Energy Harvesting
M. Carter, J. Stetter, J. Smith, A. Parks, Y. Zhao, M. Findlay, V. Patel, Proceedings
221st ElectroChemical Society Meeting, May 2012
Key enabling trend
Range scaling of far field WPT
Friis(Moore(t))
wireless power range scales exponentially (!)
but with exponent that is of EPI (IPuJ) scaling
10000000
10000000
Microprocessor Efficiency
Friis Distance (Power Limited)
Exponential Fit
Exponential Fit
1000000
100000
100000
Instructions per uJoule
10000
c2
1000
100
10000
0.46 t
1000
100
10
d2
10
0.23t
0.1
0.01
1970
0.1
0.01
1975
1980
1985
1990
1995
Year
2000
2005
2010
2015
2020
Range at which 6 K IPS workload can
be wirelessly powered (Meters)
1000000
While Inst/J is continuing to
scale, voltage scaling is
ending. Thus techniques to
efficiently harvest at low
voltage are desirable.
Inst/J
doubling time
2 years
Possible outcome if
voltage scaling limits
cannot be compensated
for
Range
doubling time
4 years
from Mapping the space of wirelessly powered systems
in Wirelessly Powered Sensor Networks and Computational RFID
Joshua R. Smith Ed., Springer 2012
What does this scaling trend mean?
Capabilities that barely work today will become robust in the future
Low end devices like RFID tags will get increasing compute
functionality, e.g. full-programmability
It will become possible to RF-power devices that seem too power-hungry
today, e.g.
microphones
cameras
Research effort is needed to overcome the voltage scaling challenge
Bad news: the further away we are, the lower the voltage
Good news: voltage is not a conserved quantitycan be boosted!
Hybrid analog / digital backscatter sensing
The original Great Seal bug aka The Thing
Our idea:
A digitally addressable passive backscatter mic
that will be read by a SW Defined RFID reader
Combine benefits of analog (low power sensing)
with digital (addressability, channel sharing, error correction, )
WISP
Amazing that RFID reader can collect sound from a totally passive microphone!
Theremin has come full circle.
Analog backscatter using SW Defined RFID Readers | Talla, Buettner, Wetherall, Smith
IEEE RFID 2012
Digital / analog backscatter traces
1 foot
Message exchanges between Reader and WISP
5 feet
Default mode: Digital mode
EPC Gen2 inventory followed by analog backscatter mode (1 second long)
After 1 second in analog mode, switch to digital mode
10
4/11/2012
WREL Wireless power
Analysis, Experimental Results,
and Range Adaptation of
Magnetically Coupled Resonators
for Wireless Power Transfer,
A.P. Sample, D.T. Meyer, J.R.
Smith, IEEE Transactions on
Industrial Electronics, Feb.
2011, vol.58, no.2, pp.544-554
Powering a Ventricular Assist
Device (VAD) with the FreeRange Resonant Electrical
Energy Delivery (FREE-D)
System, Benjamin H. Waters,
Alanson P. Sample, Pramod
Bonde, Joshua R. Smith,
Proceedings of the IEEE ,
vol.100, no.1, pp.138-49,
January 2012.
Future work:
Extreme wirelessly powered sensor systems
Fully implantable, fully wireless, perpetual
Heart pumps
Neural implants
Wirelessly powered telephones / listening devices
RF-powered & read camera
SMS-powered SMS messaging
Acknowledgement
Intel Science & Technology Center for Pervasive Computing
5 year center funded by Intel
NSF EEC 1028725 NSF Engineering
Research Center for Sensorimotor Neural
Engineering
http://istc-pc.washington.edu
Faculty Research Award
ECCS-0824265, "Realizing the internet
of things with RFID sensor networks"
