
sonable duty cycle of only 15% is sufficient for an acceptable
discovery latency. Another 36mW are dissipated by the volt-
age regulator. In total the PMM consumes less than 117mW.
The embedded Linux-system adds an amount of 5000mW
average when it is running.
Component I [mA] P [mW] Duty Cycle
uC 3 12.375 100 %
XBee 110 453.75 15 %
Voltage Reg. 3 36 100 %
Total 116.4375 mW
5. CONCLUSIONS AND FUTURE WORK
The combination of a solar charging circuit with a low
power radio for discovery and remote wakup mechanisms
into a single module benefits from synergies, e.g. only a single
MCU and a single voltage regulator is required. Therefore,
our design has lower physical dimensions, costs, and a lower
energy consumption than seperate modules for charging,
swiching, metering and discovery.
The range measurements have shown that the range of
the discovery radio exceeds 802.11a and therefore our design
works as intended. For mobile scenarios a sleeping node can
be booted up before the 802.11a contact starts. For future
work we plan the evaluation of several other discovery radios,
especially since most of the PMM’s energy consumption is
caused by the radio. To extend the range of the 802.15.4
radio, an external antenna can be used. Another possible
approach is to implement scheduled deep sleep modes with a
very low duty cycle, in which the radio listens only every few
minutes for a wakup signal. However, this would increase
the latency and might therefore not be acceptable for some
applications.
Another area of future work is the development of an
interface between the PMM control daemon and the DTN
bundle protocol agent. This allows for energy aware DTN
routing and the integration of bundle priorities into the
wakeup strategies. Moreover, since the DTN node can direcly
access the PMM’s radio, a low bandwidth convergence layer
could be implemented that uses the PMM’s radio. Another
interesting approach is that the PMM could receive and
buffer bundles while the DTN-node sleeps or is in the process
of booting up.
6. ACKNOWLEDGEMENTS
This work was supported in part by the European Regional
Development Fund under project number W2-80028895.
7. REFERENCES
[1] N. Banerjee, M. D. Corner, and B. N. Levine, “Design
and Field Experimentation of an Energy-Efficient
Architecture for DTN Throwboxes,” IEEE/ACM
Transactions on Networking, vol. 18, no. 2, pp. 554–567,
April 2010.
[2] E. Meehan and K. Hartnett, “N4c dtn node user
documentation, dtn node build,” Report, 3 2011.
[Online]. Available: http://www.n4c.eu/Download/
n4c-wp5-052-dtn-node-build-12%5B1%5D.pdf
[3] S. Farrell, “N4c dtn node design,” Report, 5 2009.
[Online]. Available: http:
//www.n4c.eu/Download/n4c-tcd-006-d5 1-FINAL.pdf
[4] S. Farrell, S. Weber, A. McMahon, E. Meehan, and
K. Hartnett, “An n4c dtn router node design,” in 1st
Extreme Workshop on Communication, Laponia,
Sweden, August 8-14 2009., aug 2009.
[5]
Digi International Inc., “XBee/XBee-PRO RF Modules,”
Datasheet, Minnetonka, USA, Sep. 2009. [Online].
Available: http:
//ftp1.digi.com/support/documentation/90000982 B.pdf
[6] M. Doering, S. Lahde, J. Morgenroth, and L. Wolf,
“Ibr-dtn: an efficient implementation for embedded
systems,” in Proceedings of the third ACM workshop on
Challenged networks, ser. CHANTS ’08. New York, NY,
USA: ACM, 2008, pp. 117–120. [Online]. Available:
http://doi.acm.org/10.1145/1409985.1410008
[7] Atmel Corporation, “AVR2016: RZRAVEN Hardware
User’s Guide,” Application Note, San Jose, USA, Apr.
2008. [Online]. Available: http://www.atmel.com/dyn/
resources/prod documents/doc8117.pdf
Comentários a estes Manuais