58 PERVASIVEcomputing Published by the IEEE Computer Society ■ 1536-1268/07/$25.00 © 2007 IEEE
Works in Progress
■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■ ■
OPEN SOURCE MOBILE-
CENTRIC WIRELESS SENSING
Dirk Trossen and Dana Pavel,
Nokia Research
The Nokia Remote Sensing platform
for mobile-centric wireless sensing ex-
ploits ubiquitous mobile devices’ intelli-
gence to connect locally to sensors and
local sensor networks. It connects geo-
graphically dispersed sensor deploy-
ments, performing both local and remote
sensing experiments with almost any
mobile phone. NORS integrates with lo-
calized sensing solutions, connecting to
a centralized server that hosts the appli-
cation functionality.
The NORS platform design
• minimizes execution costs through
event-based sensor acquisition with
local aggregation in the mobile device;
• supports future sensors through a
flexible plug-in structure;
• builds on a common data model with
the goal of supporting, for example,
TinyML;
• enables wide deployment; and
• provides a viable license model for
wide acceptance in the research com-
munity.
We chose Java Mobile Information
Device Profile 2.0 as the execution envi-
ronment, allowing for a wide range of pos-
sible gateway devices. We implemented the
server platform in C++ under Windows,
with the actual server being implemented
using the NORS APIs. The platform is
available under the LGPL (Lesser General
Public License) at http://sourcefourge.net/
projects/nors, facilitating proprietary ap-
plications while keeping the middleware
open. We’re confident that our ecosystem
will stimulate research and innovation
through open source licensing.
We implemented the platform in an
environmental-monitoring scenario with
two fixed gateways each monitoring a
facility, while a mobile phone monitors
the premises’ surroundings within a
given time (see figure 1). We use environ-
mental sensors for temperature, humid-
ity, pressure, and dew point. The fixed
gateways also monitor motion. For the
mobile gateway, we use off-the-shelf
mobile phones with Bluetooth connec-
tivity. For the fixed gateways, we use an
available cellular machine-to-machine
product. All gateways use IP to remotely
connect to our application server.
All sensors are discovered and pre-
sented to the user at the server. The user
can issue acquisition queries (including
aggregation functionality); they’re exe-
cuted locally at the gateway. So, the infor-
mation that’s transmitted to the remote
application server consists of notifica-
tions that queries have been triggered.
Upon receiving the information, the
server simply generates dynamic Web
pages of the provided information.
We learned promising lessons from this
simple deployment. The platform’s stabil-
ity is good, allowing for long-running ex-
periments using standard mobile devices.
We used mobile devices from simple Java-
only phones to high-end smart phones.
We intend to create larger testbeds in
academic collaborations. Our target is
participatory sensing scenarios that
exploit the existing base of mobile
phones. We also seek to integrate the
information gateway into efforts such
Sensor Networks,
Wearable Computing,
and Healthcare Applications
Editor: Anthony D. Joseph ■ UC Berkeley ■ adja@cs.berkeley.edu
This issue’s Works in Progress department includes five submissions covering a variety of
research topics. The first two projects are sensor network oriented: a Java MIDP-based sen-
sor network platform that’s being offered as an open source development environment
and a platform for monitoring and controlling distributed power generation systems. The
third is a wearable computing application for optimizing the circuit-training process for a
group of users. The final two projects introduce pervasive computing into medical envi-
ronments with a combined electronic and paper medical charting system and an interac-
tive, multimodal, haptic-enabled medical training simulator.
—Anthony D. Joseph
EDITOR’S INTRODUCTION