1 color=black>

XYZ: A Motion-Enabled, Power Aware Sensor Node Platform for Distributed Sensor Network Applications
XYZ: A Motion-Enabled, Power Aware Sensor Node Platform
for Distributed Sensor Network Applications
Dimitrios Lymberopoulos and Andreas Savvides
Embedded Networks and Applications Lab (ENALAB)
Yale University
51 Prospect St, Room 000
New Haven, CT 06520
email:
{dimitrios.lymberopoulos, andreas.savvides}@yale.edu
Abstract This paper describes the XYZ, a new open-source sensing
platform specically designed to support our experimental research in
mobile sensor networks. The XYZ node is designed around the OKI
ML67Q500x ARM THUMB Microprocessor and the IEEE 802.15.4
compliant CC2420 radio from Chipcon. Its new features include support
for two different CPU sleep modes and a long-term ultra low power sleep
mode for the entire node. This allows the XYZ and its peripheral boards
to transition into deep sleep for extended time intervals. To support
mobility hardware control and computations, XYZ supports a wide
dynamic range and power options. In low power conguration the node
resembles existing small low power nodes. When needed, the node can
scale up its resources to perform more powerful computations. Mobility
is enabled with an additional accessory board that allows the node to
move along a horizontal string. In this paper we provide an overview of
the XYZ architecture and provide an insightful power characterization
of the different operational modes to allow the users to optimize their
platforms for power.
I. I
NTRODUCTION
The rapid progress in sensor networks is constantly revealing new
unexplored problems and applications that create the requirement for
a diverse set of sensing platforms. On one hand, the creation of
tiered architectures using heterogeneous nodes calls for a variety of
capabilities and sensor interfaces at different levels of the hierarchy
and optimized ultra low cost sensor nodes at the leafs of the
network. On the other hand, the lack of complete understanding of
many sensing phenomena and applications suggest that the validation
of algorithms aiming to develop a fundamental understanding of
dealing with sensing uncertainties should receive a higher priority
over rigorous power optimizations. Two illustrative examples include
the recent efforts for learning-based radio signal strength (RSS)
localization [6], and the efforts to exploit mobility in sensor networks
[8]. From a research perspective,
the priority in both applications is to discover and demonstrate
distributed schemes that operate correctly and robustly on noisy
sensor data. Such an endeavor can be better facilitated with the
existence of suitable low cost and exible low cost platforms that
reduce the overhead of experimentation and preliminary deployment.
The XYZ platform takes a forward step in this direction by
instantiating a new sensor node platform designed to support mobility
experiments in sensor networks. Although our design is driven by the
research requirements of our group, extra effort was taken during the
design phase to specify a feature set that is complimentary to existing
platforms and can serve multiple aspects of research and education in
sensor networks. The XYZ platform described in this paper is built
around an ML67Q500x series ARM/Thumb microcontroller from
OKI Semiconductor and a CC2420 radio with a 250kbps raw data
rate from Chipcon. The choice of the OKI microcontroller provides a
wealth of peripherals and exible modes of operation. The Chipcon
radio and its use with an IEEE 802.15.4 compliant MAC protocol,
Fig. 1.
The XYZ sensor node. Left to right: instructional XYZ, testbed
module, XYZ in motion
make our node interoperable with other sensor nodes available in the
community such as Telos and Micaz. An additional mobility board
inspired from the outdoor NIMS system [15], allows the node to
move along horizontal strings in indoor environments. Our design is
based on the following key design considerations.
Long term sleep modes The node should be capable of long term
sleep modes, that can power-off not only the processor but also all
its sensors and peripherals for extended periods of time.
Support for mobility The node should have support for mobility.
This includes both autonomous node mobility and mobility of wear-
able nodes. In addition to having a mobility accessory board, support
for mobility also requires faster computation when the node is mobile,
and real-time support hardware (such as hardware timers, PWM
outputs and a large number of external interrupts) for controlling
motors and other actuators
Computation & Memory The node should have ample computation
resources and memory to allow researchers to quickly prototype and
test their ideas before optimizing their code.
Multiple operational modes The node should support dynamic
operation over multiple operational modes that will allow it to have
low power characteristics comparable to lower end nodes but also
provide it with the option to perform faster and more powerful
computations when needed.
A wide choice of sensing peripherals close to the sensors From
our previous experiences with sensing platforms we found out that
the ability to do fast sensing close to the main processor makes pro-
totyping easier. It eliminates several software development overheads
and bottlenecks associated with moving data between processors and
reduces the programming learning curve for new node users.
Cost and form factor The node should have a small form factor and
compact packaging wearable applications and ubiquitous deployment.
The node should also be low cost to allow the creation of scalable
testbeds for experimentation. Fig. 2.
An overview of the 3-D testebed installed in ENALAB at Yale
University.
In this paper we introduce the rst generation of the XYZ node,
our rst attempt to produce such a platform. In the next section, we
motivate the use of XYZ in the context of our 3-D, battery operated
testbed. This is followed by a description of the XYZ architecture
in section III and detailed power characterization in section IV.
Section V surveys similar platforms and section VI concludes our
presentation.
II. 3D T
ESTBED
O
VERVIEW
The main driver behind the XYZ specication is a three di-
mensional, battery operated scalable testbed under construction at
ENALAB at Yale. The testbed (described in Fig.2) is a three-
dimensional structure measuring 4.5m(W)
x
6m(L)
x
3m(H)
deployed inside our lab. It is designed to host a large number of static
and mobile nodes to instrument a variety of scenarios related to node
localization, sensor calibration and mobile sensor applications such as
boundary estimation [8]. A group of motion-enabled suspended nodes
will move horizontally and vertically along a mesh of strings. Another
group of nodes-on-wheels will roam around the testbed oor. The
sensor nodes will respond to a set of stimuli generated by articial
smoke and heat sources and a video projector mounted on the testbed
ceiling facing the oor. The consideration of the testbed created the
need for a new node architecture that is motion enabled, easy to
program and it can sustain a wide variety of sensors. To facilitate
future ad-hoc deployment and to allow a high degree of mobility
in our testbed, we require that the nodes in the testbed can operate
on batteries for a prolonged period of time. For these reasons we
designed the XYZ node to have a rich set of interfaces, a large
dynamic range of operational modes and a deep sleep mode for the
entire node. The details of the XYZ architecture are described in the
next section.
III. T
HE
XYZ A
RCHITECTURE
The XYZ architecture is depicted in Fig 3. The main innovative
hardware features are its ability to support long-term sleep modes
through an external supervisor circuit, mobility and exible pro-
cessing modes. The main processing unit is an OKI ML67Q5002,
ARM7TDMI microcontroller [10]. We found of this processor to be
an appealing choice since it provides a rich set of peripherals, multiple
power options and a suitable memory conguration. On XYZ the
processor is set to operate at a maximum CPU clock frequency of
XYZ
Operating Mode
Component
1
2
3
5
6
LDS
CPU
57.6 -
standby
halt
off
mode
1.8 MHz
I/O
Timers
Custom
none
Timers
Custom
off
Cong.
only
only
off
Radio
on / off
off
TABLE I
T
HE
XYZ
S DIFFERENT OPERATING MODES
.
57.6MHz but an internal software controlled clock divider can slow
down the processor in powers of tw