
Design and Implementation of a Low-Power Energy
Management Module with Emergency Reserve for Solar
Powered DTN-Nodes
Michael Doering, Stephan Rottmann, Lars Wolf
Institute of Operating Systems and Computer Networks
Technische Universität Braunschweig
Braunschweig, Germany
(mdoering|rottmann|wolf)@ibr.cs.tu-bs.de
ABSTRACT
DTN nodes often idly wait for contacts and unnecessarily
consume energy during these periods. For solar-powered
nodes this means that solar panels and batteries could be
much smaller with an efficient energy management, reducing
the physical size of these nodes. We present our design of a
solar-powered DTN node including a module that handles so-
lar charge management, energy management and a discovery
mechanism to wake sleeping nodes. Moreover, the system
senses the remaining battery capacity, and keeps a reserve
for emergency communications. Our evaluation shows that
the power management module’s design is energy-efficient
and performs as intended. Furthermore, we evaluate the
discovery mechanism. The results show that the capabili-
ties of the module offer a promising new approach to the
implementation of energy efficient routing in DTNs.
Keywords
Delay Tolerant Networks, Solar Power, Energy Harvesting,
Extreme Communication
1. INTRODUCTION
Photovoltaic cells are a good option for powering DTN-
nodes in remote areas which do not have sufficient infras-
tructure support. However, bundle protocol agents require
non-negligible processing power and most DTN-nodes usually
require a high bandwidth long range wireless interface. This
results in a typical power consumption in the order of a few
Watts and requires a certain area of solar panel and a certain
rechargeable battery capacity. In other words: solar-powered
DTN-nodes with a decent performance are large and bulky.
In order to reduce the physical dimensions it is necessary
to reduce the power consumption by implementing power
management. A common approach is to power-off unneeded
components and to reduce the chip frequency of underutilized
resources.
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. To copy otherwise, to
republish, to post on servers or to redistribute to lists, requires prior specific
permission and/or a fee.
ExtremeCom2011, September 26-30, 2011, Manaus, Brazil.
Copyright 2011 ACM XXX-X-XXXX-XXXX-X/11/09 ...$10.00.
Our approach goes a step further: we propose a dedicated
power management module (PMM) to switch the complete
DTN-node off. Only the PMM remains active (but in a very
deep sleep mode) and wakes the node e.g., on contact with
another specific node. The PMM also features a low-power
IEEE 802.4.15 radio for discovery and to communicate with
other nodes’ PMMs in order to negotiate complex wakeup
rules, e.g., ’wake only if battery is above 30% or if bundle
is urgent’. Moreover, the PMM handles battery charging
and accounting, and keeps an emergency reserve for urgent
communication.
In this paper, we present our solar-powered DTN-node, the
PMM hardware design and its firmware functions. Besides
the obvious advantages of power management, our system is
also beneficial for an easy implementation of energy-aware
opportunistic routing, since a DTN-router can upload dy-
namic wakeup-rules to the PMM before the node is powered
off. Our vision is a DTN in which the majority of nodes are
turned off while their batteries are charging, instead of being
drained by a node that idly waits for a contact.
Our approach is similar to the dual-radio two-tier archi-
tecture presented in [1], but there are several significant
differences. First, we use IEEE 802.15.4 for discovering other
nodes instead of a proprietary radio. This standardized wire-
less technology makes our design more inter-operable since
there is a variety of radios available (e.g. USB dongles) and
it can be expected that 802.15.4 will be integrated in future
smartphones and laptops. Moreover, the power consumption
in receive mode is lower, making discovery in ’deep sleep’
mode more energy efficient. Second, instead of using mobility
prediction, our approach uses selective discovery based on
DTN routing decisions, i.e., the PMMs actively negotiate
wakeup events based on routing and remaining battery ca-
pacity. This allows to reserve a certain capacity for urgent
or emergency communications. And third, our DTN-node is
deliberately designed to be as simple and robust as possible.
In that regard our design is similar to the solar-powered
node developed in the N4C project
1
described in [2, 3, 4].
We also decided for sealed lead acid batteries, because these
a more robust and safe than LiPo-batteries, although the
energy-density is lower. However, lead acid batteries are
much easier to obtain and available even in very remote
areas - an important requirement in developing regions. But
besides using solar cells and the same type of batteries to
power a DTN node, our design is very different from the N4C
1
http://www.n4c.eu/
Commentaires sur ces manuels