You are on page 1of 44

Special Theme: The Sensor Web

Introduction to the Special Theme

The Sensor Web: Bridging the Physical-Digital Divide

by Mark Roantree and Mikko Sallinen

One of the truly multidisciplinary research efforts involving toring needs to be continuous, integrated and without loss of
computer scientists revolves around the topic of sensor net- data, requiring the specification and deployment of software
works. It brings together chemists who develop the sensors, services for the Sensor Web. Personal health (or pHealth) net-
engineers focusing on wireless platforms and other hardware works are emerging in many research projects and industrial
components, and the computer scientists who develop the applications. Wearable sensors transmit a variety of sensed
services, knowledge layers and middleware. In many cases, readings from human participants, which are harvested and
research must also include the knowledge workers associ- undergo semantic interpretation to allow domain specialists
ated with the specific domain, many of whom are repre- to make informed decisions on the health and increasingly the
sented in the articles in this issue of ERCIM News. In almost performance of individuals in sporting environments. There
all cases, some aspect of the research will seek to create a are several commercial products in this field that enable
bridge or bidirectional channel between the physical world researchers and companies to develop more advanced solu-
of the planet, its people and the sensors, and the digital world tions for the market. Applications of the Sensor Web covered
of computers and their software applications. in this issue include exploration in oil and gas fields, multi-
media sensing, life-logging of human actions and interaction,
The emergence of the Sensor Web concept is due to the pro- and environmental hazards, demonstrating the multi-discipli-
liferation of physical devices that are accessible through the narity of Sensor Web research and highlighting the need to
internet and thus, act as an extension to the World Wide Web. bring expertise from different backgrounds together.
Through new hardware peripherals, connected directly to the
Web, automatically interpreted, integrated and transformed So what issues arise from the articles presented here? There
for human interaction, querying and mining, we create the is evidence of a large number of sensor networks in different
Sensor Web. disciplines, as already discussed. They incorporate both small
cheap devices and larger customized, proprietary and highly
The Sensor Web provides a platform for new ideas and expensive devices. In general, the sole difference between
applications for different domains. However, each applica- them is accuracy. As the smaller, less accurate sensors
tion domain has its own unique characteristics and the con- become cheaper, they will quickly represent the significant
cept of a general platform can be developed only for labora- majority of this device type on the Sensor Web. In many
tory tests. As a result, development tends to focus on propri- cases, more than one sensor will be required to support deci-
etary solutions to meet a varied set of requirements. sion-making processes. This will demand synchronization
and normalization of sensor feeds before integration takes
The breadth of research in the Sensor Web domain is place. While this presents problems, as highlighted in a num-
demonstrated in the articles in this issue of ERCIM News. ber of the articles, the power of the Sensor Web is that it pro-
Wireless networks are necessary to connect to sensor vides an infrastructure for harvesting the data. Historically,
devices that may be physically unreachable. While sensor significant volumes of data generated by sensing devices
data is often analysed after sensing has stopped, wireless have been lost, mainly due to a lack of computer scientists in
networks are essential if we are to perform live queries of the research project. This illustrates the gap between the
sensor output, and adapt the behaviour of the sensor in real physical and digital worlds.
time. Toolkits for maintaining sensor networks, together
with standards for processing and managing sensor data, The first step in developing a Sensor Web system is the con-
assist in building more powerful and robust networks. In struction of a simulation for the planned sensor network. This
addition, new technology for developing and integrating process becomes easier with time as the domains and envi-
smaller nodes enables measuring devices to be placed in a ronments are better understood. However, when designing a
far wider range of products. In an increasing number of new architecture or software service for one of the layers in
applications and projects, data generated by sensor devices an architecture, or perhaps for a new domain, precise simula-
is of a confidential nature, perhaps in areas such as personal tion is of considerable help prior to implementation into real
health or body networks. This requires the appropriate components. A necessary requirement is that all layers are
research effort into security for the data transmitted by the accurately modelled, otherwise the simulation will give mis-
many sensor devices inside the networks. Ubiquitous sys- leading results. When this step is completed, the physical
tems will also provide significant data volumes and chal- process of sensor deployment and sensor network construc-
lenges for the Sensor Web. Similarly, environmental moni- tion can begin.

14 ERCIM NEWS 76 January 2009

New ERCIM Working Group on the Sensor Web
A new ERCIM Working Group on the ing), and through their collaborators have longer term. The Working Group is
"Sensor Web" was recently established. a broad multidisciplinary base. Areas of preparing for the next round of FP7 calls
Representatives from seven ERCIM mem- interest to the group include both applied and intends to participate fully with
bers (IUA, ICS-FORTH, ISTI-CNR, and basic research. Examples of deploy- ERCIM's Fellowship Programme.
CRCIM, VTT, SARIT and NTNU), partic- ment areas include personal health, envi-
ipated in the kick-off meeting on 19 May ronmental analysis, ambient intelligence, ERCIM Working Groups are open to any
2008, or expressed their interest in joining locomotive and large vehicle monitoring, researcher in the specific scientific field.
the Working Group. military applications, deployment of per- Scientists interested in participating in the
sonnel in toxic environments, and traffic ERCIM Sensor Web Working Group
Objectives analysis. should contact the coordinator.
The objectives of the Sensor Web Working
Group is to promote and facilitate interac- Future Plans Link:
tions between various R&D groups inside The group intends to establish an ERCIM
and outside ERCIM, in multidisciplinary International Conference on the Sensor
themes relevant to the Sensor Web. The Web. This should be of a high quality, with Please contact:
Working Group members cover a wide the intention of raising the group's profile, Gregory M. P. O'Hare
range of ICT skills (software engineers, providing a forum for discussing the Sensor Web Working Group chair
information management and databases meaning and scope of the Sensor Web, and The CLARITY Centre for Sensor Web
including information retrieval, wireless attracting new people for future collabora- Technologies, University College Dublin,
applications, networks, security and e- tions. The launch of a journal on Sensor Ireland
mobility, ambient and ubiquitous comput- Web has also been considered for the E-mail:

The first task for computer scientists in Sensor Web research such as SQL and XPath or XQuery. It is likely that the XML
is to ensure data crosses safely from the physical to the dig- query languages will find widespread use, as data converted
ital world, where it can be processed and manipulated to bet- to XML has highly interoperable properties. This is crucial
ter inform us at to how to proceed in the many environments when integrating sensor data, both within a single sensor net-
in which sensors now exist. In the past, environmental and work and with the data generated by other sensor networks.
personal health sensor networks have generated large vol-
umes of data that were not captured in a digital format. For The final challenge is the identification of the most impor-
example, sports scientists have for some time been running tant issues in each application, eg closing the loop, control-
sensor-based tests on athletes, in many of which data is ling parameters, devices or actuators or giving instructions.
recorded manually and on paper. Even complex sporting These tasks should all be completed by accurate miniature
equipment such as speed gates that record the velocity of sensors and nodes, wideband data communication, and the
players as they move through various sections of a training utilization of real-time control with minimum or zero power
course, will have their data transmitted wirelessly to a hand- consumption. This final step demonstrates the path from the
held device, which then has no means of transferring the data original sensor device through the engineering layers
to a persistent storage mechanism. Eventually, sensor hard- required for transmission of data, through the software serv-
ware will always provide a means of recording and transfer- ices and human interaction, and finally back to the sensor,
ring data but in the meantime, it is the role of data manage- where the knowledge generated is used to make the sensing
ment researchers to devise a means for ensuring that this device more powerful and more accurate. The aspiration for
data is recorded electronically and stored in persistent, the Sensor Web is that it should continue to evolve and
query-capable systems. address its limitations, so that the outcomes of the sensor age
lead to an improvement in the planet's environment and the
Analysis of the data and uncovering the essential issues from health of its citizens.
huge volumes of information is the next step. If data remains
in the raw format generated by sensors, many knowledge
workers will be unable to express the complex queries that Please contact:
are required to extract knowledge or make the assessments Mark Roantree, Dublin City University, Ireland
required to adapt the behaviour of sensors within the net- E-mail:
work. The next challenge for computer scientists is to con-
vert the raw data into a usable format, preferably one that Mikko Sallinen, VTT, Finland
can be queried and updated by standard query languages E-mail:

ERCIM NEWS 76 January 2009 15

Special Theme: The Sensor Web

Sensor Networks in the Real World

by Steven D. Glaser and Tommi Parkkila

At the Center for Information Technology Research in the Interest of Society, University of California,
Berkeley, USA, we have been developing and deploying wireless sensor systems for ten years. We
have focused on solutions to societal needs. As we ask more and more from our 'motes' and their
low-power networks, we foresee important applications for sensor and control networks that will
require a more powerful and flexible solution.

The system we propose, AdapSys, is proliferation of wireless sensor net- four motes were distributed over the
based on a fundamental unit that can works, and these have completely main span and southern tower (see Fig-
perform at a very high level of abstrac- changed what we can measure. Each ure 1), comprising the largest wireless
tion - a multi-level controller and sensor element of the network is commonly vibration sensor network ever installed
hub that is completely software recon- called a 'mote' or smart sensor. Motes for structural health monitoring pur-
figurable, including basic and ancillary are combined into large networks that poses. The spatially dense array
functionality. In this scheme, each unit allow dense and detailed sensing. These resulted in an increase in effective sig-
can act as a single complex controller as networks move beyond the idea of a nal-to-noise ratio compared to single,
part of a locally controlled mesh, which sensor as a single instrument measuring isolated, sensors, and most importantly
in turn can be part of a wider distributed one thing, to a comprehensive system allowed the higher modes, both vertical
or hierarchical control network. All ele- consisting of many small nodes work- and torsional, to be analyzed easily and
ments of this system consist of the same ing cooperatively. Engineering and sci- accurately.
hardware, but have fundamentally fluid ence, however, remain captive to the
behaviors based on software adaptivity traditional hierarchical embedded sys- Deep Underground Science and
and reconfigurability. tem. This experience has led us to Engineering Laboratory (DUSEL)
devise a new monitoring and control DUSEL is a large physics and engineer-
We want to know what our structures appliance, each interacting in an ing laboratory being constructed in the
are doing: structures in the big sense, organic network. old Homestake gold mine in Lead, SD,
from our bodies up to large industrial USA. We are developing a deep in situ
processes, airframes and buildings. This Here is an example of the current state seismic observatory that will move us
has traditionally been a troublesome and of practice. During 2006 a mote net- closer to the realization of rapid imag-
expensive problem. Recent improve- work was designed, implemented, ing of dynamical geo-processes at
ments in sensors based on Micro-Elec- deployed and tested on the Golden Gate depth. More than 12 000 small-diame-
tro-Mechanical Systems (MEMS) and Bridge in San Francisco, in order to ter (~ 65 mm) exploration holes exist
in wireless technology have allowed the monitor its structural condition. Sixty- throughout the mine, which we intend

Figure 1: Mote antenna for

the accelerometer package at
the top of the South tower of
the Golden Gate bridge.

16 ERCIM NEWS 76 January 2009

Figure 2: The fundamental system unit of the AdapSys system., which includes options to act as a wireless sensor network.

to use as multi-point monitoring probes. demand through sensing, control and sonde is now software on the FPGA,
Any motion in the rock mass is thus computation. AdapSys is a compound there is little need for upgrades to entail
surrounded by multiple receivers, of flexible, reconfigurable, FPGA- physical hardware replacement; a com-
which greatly constrains the inversion based fundamental system units (see pletely new set of machines can be
back to source movements. This solu- Figure 2). One of these units can exe- implemented by installing new software
tion led us to propose the AdapSys cute the functionality of several micro- over the Web.
appliance. controllers through its multi-processor
capabilities. New functionalities can be The AdapSys prototype is currently
AdapSys added to the system as parallel self-con- being assembled at the VTT Laboratory
AdapSys is an elegant, straightforward, tained processor units inside the single in Oulu. We are planning a joint
flexible and reconfigurable system FPGA chip. This allows the system to research project with companies from
comprised of Field Programmable Gate be incrementally upgraded in the field the machine and automation industry in
Array (FPGA)-based units. Each Adap- while allowing support of modular ver- Finland, in order to test and refine the
Sys unit is: (i) a real-time multi-channel ification and certification. system in actual field situations.
data acquisition platform; (ii) a multi-
sensor data aggregator; (iii) a remotely Analyses of the DUSEL results along
reprogrammable multilevel controller; with past experience show that
(iv) highly portable; (v) distributed; and improved seismic arrays can be con- Please contact:
(vi) an embedded sensing and control structed from a string of accelerometer Steven D. Glaser,
network solution. We envision a group pods installed along a bore hole. We are Center for Information Technology
of AdapSys units controlling, say, a currently prototyping down-hole son- Research in the Interest of Society,
large paper mill, public conveyance des based on an Altera FPGA. The University of California, Berkeley,
systems, public safety equipment dur- device has 24 input channels, both ana- USA
ing a natural or man-made disaster, or log and digital, with a virtual real-time E-mail:
even an array of wind generators with machine for each. Within the FPGA
built-in nondestructive evaluation sys- there are also real-time machines for the
tems. real-time clock, bus handling, and Tommi Parkkila
numerous control loops. All memory VTT - Technical Research Centre
AdapSys uses a single FPGA to orches- functions are handled seamlessly within of Finland
trate and carry out the application the FPGA. Because the heart of the E-mail:

ERCIM NEWS 76 January 2009 17

Special Theme: The Sensor Web

MARWIS: A Management Platform

for Heterogeneous Wireless Sensor Networks
by Gerald Wagenknecht, Markus Anwander and Torsten Braun

A heterogeneous wireless sensor network (WSN) contains different types of sensor nodes. To
operate such a WSN, we present MARWIS (Management ARchitecture for WIreless Sensor
Networks). It uses a wireless mesh network as a backbone and offers mechanisms for visualization,
monitoring, reconfiguration and updating program code.

A wireless sensor network (WSN) may wireless mesh network (WMN) oper- platform can be easily integrated into
run different applications for different ates as the backbone and builds the the heterogeneous WSN.
tasks, such as event detection, localiza- communication gateway between these
tion, tracking, or monitoring. Different sensor sub-networks, the WSN and the The architecture used to manage hetero-
types of sensor node are therefore Internet. Wireless mesh nodes perform geneous WSNs efficiently contains the
required, and to handle heterogeneous the management tasks, and are con- following structural elements: one or
WSNs with a large number of these dif- trolled by a management station located more management stations, several
ferent sensor nodes, a comprehensive in the Internet. A possible scenario is mesh nodes as management nodes, sen-
management architecture is also neces- shown in Figure 1. sor node gateways plugged into a wire-
sary. We present MARWIS, a Manage- less mesh node, and the heterogeneous
ment Architecture for heterogeneous The use of a hierarchical architecture sensor nodes. The management func-
Wireless Sensor Networks, which sup- has various advantages. Sensor nodes, tionality is placed on the wireless mesh
ports common management tasks such which are normally unable to communi- nodes, meaning the resource-limited
as visualization, monitoring, (re)config- cate with each other due to incompatible sensor nodes have fewer management
uration, updating and reprogramming. It radio chips, can be interconnected using functions to perform, which in turn
takes into account the specific character- wireless mesh nodes. Furthermore, reduces memory and computation
istics of WSNs and the restricted physi- dividing a huge WSN into smaller sen- requirements. A user can perform man-
cal resources of the sensor nodes. These sor sub-networks decreases the number agement tasks using a management sta-
include battery life, computing power, of hops required to reach each sensor tion, and this can be remotely located
memory, network bandwidth and link node. Specifically, each sensor node on the Internet.
quality. reaches the next wireless mesh node
(which is the communication gateway) Using a graphical user interface, the
One of the main features of MARWIS is within three to four hops. This results in topology of the heterogeneous WSN
its hierarchical architecture. We divide a better communication performance with with all the sensor sub-networks is visu-
large heterogeneous WSN into smaller a lower round-trip time, lower jitter and alized. The status information about
sub-networks, each of which contains less packet loss. A further advantage of every sensor node is monitored and dis-
sensor nodes of one specific type. A using a WMN is that a new sensor node played. This includes hardware features

Figure 1: A possible MARWIS scenario.

18 ERCIM NEWS 76 January 2009

Figure 2: The WSN manager provides the
management functionality for the different
sensor sub-networks. It consists of three
databases and the MARWIS server with
three modules.

(micro-controller, memory, trans- and energy. If newer information is

ceiver), software details (operating sys- required, the sensor node can be
tem versions, protocols, applications), queried directly.
dynamic properties (battery, free mem-
ory) and, if available, geographical The MARWIS server contains three
position information. The applications modules for the management tasks. The
running on the sensor nodes or network WSN monitor module connects to the
properties can be reconfigured using the WSN information database and to the
user interface. Furthermore, updating sensor value database in order to handle
and reprogramming the sensor nodes is requests from the management station.
a very important issue. In large WSNs It also stores data coming from the sen-
manual execution of this task is unfea- sor nodes into the databases. The WSN
sible, and a mechanism to handle it configurator module is responsible for
automatically and dynamically over the the configuration tasks. It queries prop-
network is required. Both the operating erties from the sensor nodes and stores
system and applications must be them in the WSN information database.
updated, either fully or partially. The code update manager module
stores newly received program images
The WSN manager located on the mesh (and related information) in the pro-
nodes provides the management func- gram version database and notifies the
tionality for the different sensor sub- management station about available
networks. It consists of three databases programs. Link:
and the MARWIS server with three
modules, as shown in Figure 2. The Sensor Node agent is the comple- mancom.html
ment of the MARWIS server and per-
The WSN information database stores forms the management tasks on the sen- Please contact:
all information about the sensor nodes sor nodes after message exchange with Gerald Wagenknecht
and the WSN, such as the topology the MARWIS server. University of Bern, Switzerland
(neighbours, addresses) and states of Tel: +41 31 511 26 36
the sensor nodes (battery, memory). The architecture is currently being E-mail:
The program version database stores all implemented and tested in a small real-
versions of all programs for all plat- world testbed. A small Linux distribu- Markus Anwander
forms, which can be installed in the sen- tion (kernel is running on the University of Bern, Switzerland
sor nodes. Finally, the sensor value mesh nodes; the MARWIS server is Tel: +41 31 511 26 34
database stores all data measured by the being implemented in C using sockets; E-mail:
sensors. To get information about the the databases are managed with
sensor nodes, first the databases on the MySQL; the API for accessing the data- Torsten Braun
relevant mesh node are queried. This bases is implemented in C; and Contiki University of Bern, Switzerland
means a direct connection to the sensor is running on the sensor nodes as the Tel: +41 31 511 26 31
node is unnecessary, which saves time operating system. E-mail:

ERCIM NEWS 76 January 2009 19

Special Theme: The Sensor Web

The IBM Mote Runner

by Thorsten Kramp, Michael Baentsch, Thomas Eirich, Marcus Oestreicher, Ivan Romanov
and Alexandru Caraças

Wireless sensor networks may well be the next big thing. Nevertheless, a fully business-process-
integrated infrastructure for deploying large numbers of sensors and actuators requires a well-designed
ecosystem. This should combine inexpensive devices with simple, bulletproof device programmability for
easy integration and use by application domain specialists. The IBM Mote Runner system addresses this
challenge with a high-performance, low-footprint middleware platform comprising a hardware-agnostic
and language-independent virtual machine together with development and integration tooling to easily
create and manage applications for open sensor and actuator networks.

One-way, dedicated data-gathering IT Minimizing up-front investment tively using WSNs for data collection,
networks such as those underlying, for requires minimum hardware cost, and pre-processing and autonomic feed-
example, a delivery tracking system, this necessitates very efficient software back.
have shown the commercial value of running on the least expensive and most
real-time control of real-world compo- cost-effective off-the-shelf chips (or The IBM Mote Runner run-time envi-
nents. Building on this, more general- 'motes'). Minimizing subsequent invest- ronment for wireless sensor networks,
ized applications for wireless sensor ments translates into design require- currently under development at the
networks (WSNs) are becoming ments for minimum hardware interac- IBM Zurich Research Laboratory, tack-
increasingly apparent and significant in tion after mote deployment (eg for man- les these challenges in a holistic man-
ual battery change or sys- ner. At its core, Mote Runner provides a
tem reconfiguration). high-performance, resource-efficient
virtual machine that is compatible with
The second problem is high-level languages and which shields
technological: a WSN run- portable applications from hardware
time environment must not specifics. It is designed to run on very
only be able to cope with small standard embedded controllers
the broad range of techni- including low-power 8-bit processors,
cal challenges imposed on thereby reducing initial investment
WSNs but it must equally costs. Furthermore, it allows program-
be accessible beyond the mers to use object-oriented program-
low-level functionality of ming languages and development envi-
individual WSN nodes. ronments such as C# and Java to
Here, 'accessible' refers to develop portable WSN applications that
three things. First, it must may be dynamically distributed, loaded,
be possible to dynamically updated, and deleted even after the
configure and reconfigure WSN hardware has been deployed,
Figure 1: IBM Moterunner architecture. the WSN in the field to thereby reducing post-deployment and
deal with situations such maintenance costs. All operations and
as interrupted communica- communications can be cryptographi-
size and real-world relevance. Concep- tion or WSN node failures. Second, it is cally protected to establish a trusted
tually, the broadest application cate- necessary to secure the WSN in order execution environment. Figure 1 illus-
gories for WSNs involve environmental that it may be considered a trusted trates how this all fits together. Finally,
information, and provide a flexible com- source of information and reliable per- Mote Runner WSN applications pro-
munication and intelligence-gathering former of actions in response. Third, the vide seamless integration with state-of-
infrastructure that serves, for example, WSN must be well integrated into the the-art back-end infrastructures by
next-generation business applications larger infrastructure with which it coop- means of an event-driven process
by allowing them to directly tap into the erates. It must be generally programma- engine, which effectively bridges the
ever greater number of digitally-enabled ble by domain specialists to solve gap to large-scale business and scien-
sensors and actuators that provide input domain-specific problems without deep tific applications without requiring deep
to and control of their operation. knowledge of WSN technology and technology skills.
components. Only then are real-world
To unlock this potential, however, two solutions possible which link – while Link:
first-order problems must be addressed. being easy to program and deploy – the
One is cost: WSNs consist of many physical world of sensors and actuators
small computing elements that must be with business processes and applica- Please contact:
cost optimized. In this realm, cost takes tions. The result is the desired improve- Thorsten Kramp
the form of up-front investments in ment in the responsiveness of transac- IBM Zürich Research Laboratory,
hardware and software plus any subse- tions, enabling end-to-end process Switzerland
quent investments (eg for maintenance). security and reducing cost by effec- E-mail:

20 ERCIM NEWS 76 January 2009

A Testbed for Sensor Service Networks
by Thomas Usländer and Kym Watson

Working towards 'plug and measure' in sensor networks for environmental monitoring with Open
Geospatial Consortium (OGC) standards, the SANY (Sensors Anywhere) project specifies an
architecture for all kinds of fixed and moving sensors. This will allow both seamless plug-and-
measure capability for sensors in the field, and sharing of information between sensor networks.

The SANY project focuses on interoper- fications. At the sensor network level, which is able to aggregate or fuse sen-
ability of in-situ sensors and sensor net- the ad hoc wireless ZigBee network is sor data from several SOSs. The Fusion
works. This is done using both the stan- complemented by simulated sensor SOS queries the catalogue for available
dards and the on-going work of the nodes, which measure properties such SOSs of the required type and then con-
OGC (in particular the Sensor Web as temperature, humidity, illuminance ducts a selected procedure to produce a
Enablement suite of standards), OASIS and acceleration. The testbed is spatial or spatio-temporal interpolation.
(Organization for the Advancement of designed for experiments in a wide The interpolation result is a so-called
Structured Information Standards) and range of scenarios and scales, such as coverage, a function defined on a
W3C. The SANY sensor service archi- mobile sensors traversing several net- space-time grid of sampling points. The
tecture provides a quick and cost-effi- works. The simulation is implemented procedure takes the inaccuracy of the
cient way to reuse data from sensor and as an application in LabVIEW raw sensor data into account. The spa-
data sources that are currently incompat- (National Instruments), which has the tio-temporal uncertainty of the fusion
ible. Data sources can include live sen- additional task of configuring the Zig- result is specified using uncertML, an
sor data, databases of archived data and Bee nodes. New sensor nodes (either XML schema developed by the
model-based calculations. real or simulated) are recognized auto- INTAMAP (Interoperability and Auto-
matically and registered in one of three mated Mapping) project to describe the
The sensor service architecture and the OGC Sensor Observation Servers statistics of uncertain data. As with the
service specifications have been made (SOS). The sensor values are then underlying sensors, the fusion proce-
publicly available on the SANY project inserted into an SOS as they arise by dure is described with the OGC sensor
server, while the SANY specifications measurement or simulation. The avail- model language SensorML. In this way,
and best practice experience have been able network resources (observed fea- the fusion procedure can be treated as a
contributed to the OGC standardization tures, sensors, services) are registered sensor, but with the important charac-
work. The results are being tested in in a catalogue server along with meta- teristic that its result is a coverage. The
three innovative risk management appli- data to support resource discovery by coverage can be visualized using a Map
cations covering the areas of air quality, client applications. Clients can find, for & Diagram service from the SANY
marine risks and geo-hazards. example, information sources for a partner ETH Zürich.
given region and observable phenome-
The Fraunhofer Institute for Information non of interest. The procedures developed to date are
and Data Processing (IITB) has realized variants of the Bayesian Maximum
a testbed for sensors and services in Fraunhofer has produced in the testbed Entropy method that is able to consider
order to trial the architecture and speci- a special SOS known as a Fusion SOS, soft sensor data (eg where the sensor

Figure 1: Testbed for Sensor Service Networks.

ERCIM NEWS 76 January 2009 21

Special Theme: The Sensor Web

value lies in an interval) and additional ing. The intermediate files produced by three development cycles. The SANY
phenomenological knowledge on the the fusion procedure are uploaded consortium is composed of sixteen part-
relationships between observed proper- together with the fusion result to the ners from eight countries. It includes
ties. If additional sensors or SOSs enter WebGenesis information management the two research organizations Austrian
the testbed, the Fusion SOS discovers server. This ensures a reproducible trace Research Centers (coordinator of the
these new resources with the aid of the of the processing steps. consortium) and Fraunhofer, six com-
catalogue and incorporates the new data panies, three universities, four public
sources automatically into the fusion On-going work in the testbed involves authorities and the Open Geospatial
procedure. The self-describing informa- the use of RESTful Web services to pro- Consortium Europe (OGC).
tion plays an essential role in this plug- vide representations of the network
and-measure capability. resources and the development of Links:
model-based fusion methods. SANY
The Fusion SOS is implemented on the (Sensors Anywhere) is an FP6 Inte-
platform WebGenesis, an information grated Project co-funded by the Euro-
management server from Fraunhofer pean Commission within the Thematic Please contact:
IITB. The information management Priority 'Information Society Technolo- Kym Watson
server contains the information cate- gies' in the area of ICT for environmen- Fraunhofer Institute IITB, Germany
gories features of interest (sampling tal risk management. SANY is a three- Tel: +49 721 6091 486
grids), procedures and results with year project that started in September E-mail:
associated metadata to support search- 2006 and has now completed two of

Bringing the Semantic Sensor Web

to Smart Buildings
by Rob Brennan

Deploying sensor networks in the built environment is not enough to produce smart buildings. How
can we avoid creating silos of application-specific sensor networks? How can we publish the sensor
data in secure, reusable and flexible ways? Can we support end-to-end provisioning of these sensor
networks as an integral part of the building from requirements collection through design,
procurement, construction, commissioning and facilities management operations?

The Facilities Management (FM) sub- ment or space management) and techni- sor deployments and capabilities)
group of the NEMBES (Networked cal facilities management (eg energy developed by the OpenGS consortium,
Embedded Systems) project aims to management, buildings operation and and standard W3C Semantic Web tech-
answer these questions and more by maintenance). Within NEMBES-FM nology such as metadata definitions
bringing together a multidisciplinary we are extending the application of sen- encoded as RDF (resource description
team of architects, civil engineers, com- sors in smart buildings with additional framework) documents. This means
puter scientists and electronic engineers, sensing capabilities and sensor-centric adding formal semantic annotations to
who are addressing the issue of net- applications for these areas. However, existing standard Sensor Web lan-
worked embedded sensor systems in a rather than building application-spe- guages in order to provide semantic
holistic way across the stack from chip cific sensor networks and ending up descriptions and enhanced access to
design, networking, middleware and with isolated silos of non-interoperable sensor data. This is accomplished with
service management. This four-year and inflexible sensor networks, the model references to ontology concepts
project started in October 2007 and is project has adopted a semantic Sensor that provide more expressive descrip-
funded by the Irish Government's Web design to give a Web-based, open tions of and relationships between con-
Higher Education Authority under the distributed system of sensor resources cepts. The use of formal metadata to
Program for Research in Third Level within the building. This enables describe the sensors' outputs, platforms,
Institutions program. The project is led resource sharing, resource reallocation, locations and control parameters will
by the Centre for Adaptive Wireless sensor network interoperability, sensor enable a new generation of flexible
Systems, and the FM subgroup is led by discovery and intelligent applications facilities management applications to
the Informatics Research Unit for Sus- that discover and reason over associa- be built.
tainable Engineering (IRUSE). tions, for example between events in
space and time or within a particular Efficient and flexible management of
Traditionally, facilities management in context. disparate, decentralized information
the AEC (Architecture Engineering and sources such as sensor data, building
Construction) domain is concerned with This is accomplished through the appli- occupancy graphs, facilities manage-
infrastructural facilities management cation of semantic Sensor Web technol- ment process models and building
(eg security and emergency manage- ogy such as SensorML (to describe sen- information models will enable smart

22 ERCIM NEWS 76 January 2009

Figure 1: The NEMBES project.

FM applications for buildings with mul- This project also goes beyond tradi-
tiple occupying organizations and a dis- tional ICT standardization to embrace
tribution of facilities management the major AEC domain IT standards
authority across different management such as the IFC (Industry Foundation
roles. The long-term vision of the group Classes), used for describing models of
is to enable ambient intelligence within buildings in CAD tools and AEC con-
the smart building. In this scenario, struction requirements and project man-
building information models combined agement tools. Integrating our work
with location-sensing technology will with the IFC standards will enable a
allow the distribution of context-spe- dialogue with the ACE domain experts,
cific data to facilitate the monitoring of thereby encouraging rapid integration
maintenance activity progress, ie the of semantic Sensor Web technology in
'ambient interaction' of inspection and ACE tools and business practices. The
maintenance personnel with the fabric project consortium is fortunate to
of the building itself. Examples of the include the Environmental Research
benefits of this technology will include Institute at UCC, a live smart building,
assisting facilities management staff in which sensor networks and building
with automatic monitoring and support information models are combined with
of health and safety routine procedures; Semantic Web technology to provide a
avoiding illegal occupation density in unique testbed.
public spaces by triggering security per-
sonnel actions; and more effective facil-
ity space and relocation management
via automatic inventory item tracking.
Smart infrastructural FM applications Links:
that easily integrate new personnel, The NEMBES project website:
inventory items, sensors and use-cases
without the intervention of dedicated IT
personnel, will instead have these tasks The Knowledge and Data Engineering
performed by FM personnel and their Group, Trinity College Dublin:
delegates within the organizations
occupying the building. They will addi-
tionally support seamless process inte- The Centre for Adaptive Wireless
gration between local FM activities and Systems, Cork Institute of Technology: Please contact:
external third parties such as contract Rob Brennan
maintenance engineers, visitors, short- Knowledge and Data Engineering
term occupants (eg conference organiz- Informatics Research Unit for Group, Trinity College Dublin, Ireland
ers) and security or emergency response Sustainable Engineering (IRUSE) Tel: +353 1 896 8426
teams. E-mail:

ERCIM NEWS 76 January 2009 23

Special Theme: The Sensor Web

Building the Sensor Web

– Standard by Standard
by Andrew Woolf

An explosion in the instrumentation of our environment using sensors of all descriptions is driving
the development of infrastructure to manage the wealth of information they collect. The Sensor Web
aims to simplify the publication of, and access to, sensor resources, just as the World Wide Web has
done for documents. And, as with the WWW, the Sensor Web relies on new information and
communication standards for structuring sensor information and its exchange.

At the heart of these new standards is a extended to specific sensor applications the Sensor Observation Service (SOS,
conceptual model for Observations and (see box). [OGC 06-009r6]) provides a Web serv-
Measurements (O&M, [OGC 07-022r1, ice interface for retrieving filtered
OGC 07-002r3]). It says simply that an The abstract O&M model may be observations or related information
Observation is an action whose Result is applied across the spectrum of sensor (feature-of-interest, sensor parameters,
an estimate of the value of some Prop- applications and deployments, and pro- observation results). Individual sensor
erty of a Feature-of-interest, obtained vides a framework for building observations may be aggregated within
using a specified Procedure (Figure 1: exchange standards and service inter- one service into combined 'observation
The 'Observations and Measurements' faces for accessing sensor data and con- offerings' and multiple services may be
conceptual model.). Each of these core textual information (Figure 2: The Sen- federated into single access points. The
O&M information classes may be sor Web standards stack.). For example, Sensor Model Language (SensorML,
[OGC 07-000]) is an XML language for
describing observation procedures and
sensor types. Other related standards
include the Transducer Markup Lan-
guage (TML, [OGC 06-010r6]) for
transducers and transducer systems, the
Sensor Planning Service (SPS, [OGC
07-014r3]) for tasking and scheduling
observation requests with sensor sys-
tems (eg by satellite remote-sensing
instruments), and the Sensor Alert Ser-
vice (SAS, [OGC 06-028r3]) for setting
up notification subscriptions for spe-
cific sensor events.

These new standards are being devel-

oped by the Open Geospatial Consor-
tium (OGC), a non-profit de facto inter-
national standards body for geographic
Figure 1: The 'Observations and Measurements' conceptual model. information. OGC members span gov-
ernment, academia and industry. Orga-
nizations contributing to the O&M
specification include CSIRO (AU) as
editor, Geoscience Australia (AU), Uni-
versity of Alabama in Huntsville (US),
Image Matters LLC (US), Washington
University (US), Science and Technol-
ogy Facilities Council (UK), SeiCorp
Inc. (US), Galdos Systems Inc. (CA),
Geospatial Research & Consulting
(DE), PCI Geomatics (CA), and Texas
A&M University (US).

Research undertaken through global

collaborations of experts, often in the
context of coordinated international
engineering testbed activities (the OGC
Web Service Initiatives) accelerates the
Figure 2: Figure Two: “The Sensor Web standards stack. development of interoperability stan-

24 ERCIM NEWS 76 January 2009

dards and protocols. With sufficient
maturity, these specifications will be
Example: Marine Science Observation
proposed for de jure standardization
through the International Organization
for Standardization (ISO).

The O&M conceptual model has been

substantially developed since 2002
through such initiatives (OWS-1.2 and
OWS-3). It was finally approved in
2008 for Version One publication by
OGC, and is being progressed as a new
ISO standard 19156 (see links below).

A growing awareness of the impor-

tance of environmental monitoring for
the health of our planet is leading to the
development of large-scale infrastruc-
tures that transcend national bound- Figure 3: O&M example – marine research cruise.
aries. Through a ten-year implementa-
tion plan, the Group on Earth Observa-
tions (a grouping of 76 national gov- A cruise ship measures seawater temperature (O&M observed Property) along
ernments and other international organ- a North Atlantic cruise track (O&M Feature-of-interest) using a thermosalino-
izations) aims to integrate existing graph (O&M Procedure), and produces a series of data values {19.2°C, 18.7°C,
observation networks into a Global ...} (O&M Observation result).
Earth Observation System of Systems
(GEOSS) to achieve comprehensive, The SOS GetObservation operation is used to retrieve the entire observation
coordinated and sustained observation and its context; DescribeSensor will return thermosalinograph details; GetFea-
of the Earth system. tureOfInterest provides details of the cruise track; while GetResult returns just
numerical temperature values.
Within Europe, a recent Directive
(2007/2/EC) will establish the 'Infra-
structure for Spatial Information in
Europe' (INSPIRE) to integrate envi-
ronmental data across all member
states. The 'Kopernikus' partnership
between the European Commission
and the European Space Agency will point, while service standards like SOS emerge practical demonstrations of
establish core operational services (eg and SPS provide the computational their effectiveness. Demonstrators like
ocean forecasting, landcover monitor- viewpoint. the US-based OpenIOOS testbed for an
ing, emergency response) for the global Integrated Ocean Observing System
environment and civil security. These Within the broader context of ICT inno- and projects like the European 'Sensors
three global-scale initiatives all require vation, Sensor Web standards build on Anywhere' (SANY) FP6 Integrated
standard information models and net- key W3C specifications (XML, Web Project are validating the application of
work services for integrating sensor services). There is a growing move these standards in real Sensor Web
data – both in situ and remotely sensed towards richer semantics, for describing applications.
(spaceborne and airborne). observed properties and their relation-
ships; for instance, ISO 19150 is a new Links:
An underlying abstract architectural standard for the use of ontologies with
approach is used to develop Sensor Web geographic information. An important
standards. The Reference Model for principle is the decomposition of
Open Distributed Processing (RM- domains of governance – identifying markets-technologies/swe
ODP, [ISO/IEC 10746]) factors a dis- responsible parties with a remit for
tributed system like a Sensor Web into managing agreed vocabularies and con-
five complementary viewpoints: enter- cepts on behalf of a community of inter-
prise (roles, scope and policies of the est (eg definitions of particular sensor
system), information (semantics of systems and observables). This is
information and information process- required in order to facilitate reuse of Please contact:
ing), computational (service interfaces), models and enhance interoperability. Andrew Woolf
engineering (component distribution e-Science Centre, STFC Rutherford
across nodes) and technology (imple- While much of the Sensor Web stan- Appleton Laboratory, UK
mentation choices). Against this model, dardization work has so far been devel- Tel: +44 1235 778027
O&M provides the information view- opmental, there are beginning to E-mail:

ERCIM NEWS 76 January 2009 25

Special Theme: The Sensor Web

Revolutionising Sensor Based Automation

in Manufacturing
by R Harrison, F Jammes, H Smit and T Kirkham

Increased access to device-level automation components is closing the final gap in the enterprise
computing model.

The implementation of Web service- manufacturing plants and remove con- trated how the production data could be
enabled sensors and actuators on pro- trol from vendors back to users. combined with other enterprise data to
duction lines will permanently change improve the accuracy of decisions relat-
the way in which future automation sys- Implementation ing to production routing and supply
tems are designed and implemented. The SOCRADES and SODA (Service- chain management.
Current interfaces to automation com- Oriented Device and Delivery Architec-
ponents are largely vendor-specific, ture) projects have conducted research During the demonstration, data from
restricting the reconfiguration of lines and trials into automation based on the sensors and actuators on the produc-
and the management of line data across service-oriented architecture (SOA), tion line was transmitted by equipment
enterprises. Outside the automation and in the past year have delivered ini- created by Schneider Electric. This
domain, enterprise system development tial prototypes. At the recent ITEA equipment consisted of Web-service-
has seen real-time data linkage take exhibition in Rotterdam, a test rig enabled Field Terminal Blocks (FTBs),
great steps in the office, warehouse and developed with Ford was used to which support the Device Profile for
supply chain. Research into the use of demonstrate SOA-based sensor and Web Services (DPWS) toolkit. The
Web-service-enabled sensors and actua- actuator data being used to manage DPWS toolkit is designed for embed-
tors has the potential to present an open monitoring and control applications. ded systems and has a small memory
standards-based method to integrate The data was fed to project partners in footprint, but also contains a selection
production lines into this enterprise the Enterprise computing field (SAP) of Web service standards to suit the
computing model, an innovation that and industrial automation sector (ARC demands of an automation environ-
will revolutionize automation in future Informatique). The demonstrator illus- ment. Both projects are working on

Figure 1: Vision of service-enabled


SCADA SAP Human Machine



Mode Status
FTB Services

Element Each
Each Element
Logic and

Rig Sensors and Actuators

Figure 2: ITEA exhibition architecture and rig picture showing FTB location.

26 ERCIM NEWS 76 January 2009

these demands, which are focused on reducing costs by improving resource development of devices such as the
execution timing and reliable/efficient usage and reducing downtime. The FTB, pioneered in the SOCRADES and
message delivery. The FTB is a piece of greater accuracy and real-time access to SODA projects to support this new
hardware that contains an ARM 9 chip product-level data will further enhance vision.
developed to support the DPWS toolkit. enterprises by allowing them to make
more accurate decisions regarding pro-
The support of DPWS on the FTB duction and supply chain matters. The
allows Web service interfaces to be application of the results from these
developed to the device-level I/O research projects in real manufacturing
within the line. For example, calls to environments will be the subject of Links:
directly command an actuator or moni- future work that should confirm these SOCRADES:
tor specific sensors on lines can be findings. The adoption of the approach
made via Web services located on a will also be dependent on further SODA:
variety of applications, as opposed to research in the areas of safety, security
specific vendor control devices. The and real-time execution of devices. Device Profile for Web Services home
control of devices by Web services has page including link to specification:
been achieved using central Web serv- Direct SOA linkage to sensors and actu-
ice orchestration and also on a smaller- ators moves a traditionally vendor-spe- 2/devprof/
scale peer-to-peer choreography. cific computing area into a new open
domain, ready to link with existing Please contact:
Future innovations in enterprise computing. Tom Kirkham
Device-level automation components For the manufacturer this will improve Loughborough University,
that produce data in standard and open performance and reduce costs. How- Leicestershire, UK
forms will be faster to reconfigure, ever, this is dependent on the continued E-mail:

Security and Robustness

of Wireless Sensor Networks
by Václav Matyáš and Petr Švenda

Researchers at Masaryk University, Brno, are working on security issues relating to large-scale,
highly distributed and relatively dense wireless sensor networks.

In our work we focus on link key estab- While the resiliency of probabilistic nificantly better fraction of secure
lishment in the memory- and computa- pre-distribution schemes generally links than previously published SA
tion-restricted environment of wireless increases when more keys can be put protocols, especially for denser net-
sensor networks (WSNs). We also into a key ring on every single node, works. We applied SA protocols of
study how link security behaves under such an increase is limited by the node partially compromised networks
a selected attack and what methods can storage capacity. Our multiparty proto- resulting from node capture when
be used to strengthen the resilience of col creates a large virtual key ring in an probabilistic key pre-distribution are
WSNs against compromise. We base efficient and secure way from the key used, and provided analytical and sim-
our work on the assumption that a par- rings of separate nodes. This results in a ulation evidence that SA protocols
tial compromise in WSNs is inevitable substantial increase in resilience of the work even better here. On average, SA
and network architecture should be pre- underlying probabilistic key pre-distri- protocols secure more links for proba-
pared to cope with related security bution scheme against the threat of bilistic pre-distribution than for key
issues. We work with two basic link node capturing. The protocol performs infection, when networks with the
key establishment concepts based on similarly to the hypercube pre-distribu- same percentage of initially compro-
symmetric cryptography: memory-effi- tion (Liu & Ning, 2003) but is more mised links are assumed. When the SA
cient probabilistic pre-distributions suitable for scenarios with random protocols are applied, a network with
(Eschenauer & Gligor, 2002) and light- deployment and unknown link compro- half of its links compromised can be
weight key exchange without pre-dis- mise status. The proposed protocol made reasonably secure with less than
tributed secrets (Anderson et al, 2004). itself is also resilient against partial 10% of compromised links.
These two key distribution concepts compromise inside a group of support-
behave differently when the network is ing neighbours. Some combinations of SA protocols
attacked. Analysis of the resulting com- that worked for key infection do not
promised patterns has led to the pro- Our former work exploited non-unifor- increase the number of secure links in
posal of mechanisms for improving the mity of link compromise patterns in probabilistic pre-distribution and thus
network resiliency based on support key infection, and led to a secrecy only impose unnecessary communica-
from neighbouring nodes. amplification (SA) protocol with a sig- tions overhead. Instead of analysing

ERCIM NEWS 76 January 2009 27

Special Theme: The Sensor Web

Figure 1: Automatic generation of

secrecy amplification protocols.

each separate compromise pattern aris- strategies with demonstrative applica-

ing from the combination of a particu- tions to link key security for probabilis-
lar key distribution method and tic pre-distribution and key infection
attacker strategy, we proposed an auto- approach. New attacks are generated
mated approach based on the combina- either as a recombination of existing
tion of a protocol generator and net- attacks or as completely novel attacks
work simulator. We utilize evolution- automatically assembled from elemen-
ary algorithms to facilitate guided tary attacker actions. They are then
searches for high-performance SA pro- evaluated on a network simulator or in
tocols created as a series of elementary a real system. Attacker strategies that
instructions. Every candidate protocol increase the number of compromised
is evaluated on our network simulator links with respect to several determinis-
for a particular compromise pattern tic algorithms or random cases were
(see Figure 1). found. Initial results for attacks against
selected routing protocols show good
Using this method, we were able to prospects for an automated search for
automatically re-invent all the human- selective jamming, message dropping
designed SA protocols of which we and neighbour overloading to achieve a
were aware, and to find a new protocol specific attacker goal such as increasing
that outperforms these. Moreover, we routing path length, message latency or
proposed an alternative construction of concentrating routed messages.
SA protocols that exhibits only a linear
(instead of exponential) increase in nec- Due to battery power limits and taking
essary messages when the number of into consideration the high communica-
neighbours in the communication range tions overhead exposed by current
(network density) is growing, and we replication detection, the reputation
achieved comparable performance to management mechanisms that have
protocols with original message-expen- been proposed so far are often not
sive assumptions providing energy-effi- affordable. We are currently designing
cient SA protocols. With respect to clas- prevention, detection and reaction tech-
sical human-made protocols, an niques for the network. Rather than Link:
increase in the number of secure links aiming for perfect security, which is A technical report covering some of
was obtained by an efficient combina- particularly hard to achieve in WSNs, the issues discussed above:
tion of the simpler protocols and an the aim is to force attackers to make
unconventional interleaving of elemen- disadvantageous trade-offs in terms of 2007/FIMU-RS-2007-05.pdf
tary instructions. These allow protocols computational time, energy or other
to be executed even when one of the costs. Due to the diversity of usage sce- Please contact:
participants is out of radio transmission narios, there is a need to develop an Václav Matyáš
range. economical/mathematical model that Masaryk University, Brno – CRCIM,
would help to find a near-optimal solu- Czech Republic
Our current work focuses on the con- tion for a particular combination of net- Tel: +420 549 49 5165
cept of automatic search for attack work usage and available resources. E-mail:

28 ERCIM NEWS 76 January 2009

Security Challenges for Wireless Sensor Networks
– Dynamic Routing as a Security Paradigm
by Marek Klonowski, Michał Koza and Mirosław Kutyłowski

Recent work carried out at Wroclaw University of Technology shows that a fair level of security can be
achieved for wireless sensor networks without heavy cryptographic technology.

Wireless sensor networks processing

sensitive data are facing the risks of data
manipulation, data fraud and sensor
destruction or replacement. This con-
cerns applications such as the gathering
of data on environmental pollution
around industrial installations, or sensor
systems replacing traditional video
monitoring. Large-scale deployment in
practice is conditioned by solving these
kinds of security problem and reducing
the risks due to limited physical protec-
tion of the devices and openness of the
wireless communication channel. While
modern cryptography and computer
security offer many ways of solving
these problems, they are focused on
solutions for high-performance devices,
and not for computationally weak sen-
sors with limited communication band-
width. New 'lightweight' solutions tai-
lored for the special needs of wireless
sensor networks have to be designed. Figure 1: Snapshot of a single routing path.
This is one of the focal points of the EU
project FRONTS (Foundations of Adap-
tive Networked Societies of Tiny Arte-
facts). Fortunately, some recent devel- vantage that an adversary can attack the scheme has the following basic proper-
opments have shown that without heavy network by gaining control over inter- ties when processing a message M:
cryptographic technology it is still pos- mediate sensor nodes. The cryptography • Pi+1 receives encrypted messages
sible to achieve a fair level of security in used by such devices is usually weak from Pi and Ri in order to compute its
a practical sense. This report indicates a and can provide opportunities to reveal share of message M,
few ideas of this kind. information sent or to manipulate them. • Ri+1 receives different encrypted
messages from Pi and Ri in order to
Due to the energy required for transmis- The following idea may be applied in compute its share of M.
sion over long distances, it is often a order to make it much more difficult to
good idea to route data along a sensor carry out attacks. Instead of a single The encryption scheme guarantees that
network by making many hops over information path, each message is sent corrupting either Pi or Ri reveals no
small distances instead of a direct trans- over a double path. This means that information about M. Also, combining
mission from a sensor to the sink node. instead of a single ith node Ni we have the shares from different stages of mes-
However, such a solution has the disad- two nodes: Pi and Ri. The encryption sage processing gives no information

Figure 2: Encoding details.

ERCIM NEWS 76 January 2009 29

Special Theme: The Sensor Web

about M as long as the adversary has collect and corrupt many sensors until view of hiding the transmission routes
only one share from each level of the the matching pair is found. in the case of heavy traffic, under the
path. assumption that an adversary can select
Another step of the design is to make the traffic coming out of each node.
What is the advantage of such a design? the path self-evolving: at any time a This involves studies concerning com-
The main point is that while it might be node may negotiate with its predeces- binatorial issues of traffic analysis as
relatively easy to find and corrupt one sors and successors a change of the well as stochastic investigations of the
of the nodes (say Pi ) for this to be use- transmission key and redirect its duties rapid mixing of Markov chains. Further
ful, the adversary must still find and to another node. Since these changes details of the scheme will be developed
corrupt the matching node Ri . This can can be made independently and uni- in cooperation with other partners of the
be difficult for purely practical reasons: formly at random, the data path may project; in particular, we plan to
if each sensor is hidden in the environ- evolve so fast as to make unfeasible any develop a prototype of the system.
ment, then while the first might be attempt at data analysis based on moni-
found by chance, the second must be toring radio traffic. Indeed, a cryptana- Link:
found by a detailed search in the same lytic attack would face the difficulty FRONTS:
area. This could be hard without arous- that assigning the messages to sensor-
ing the interest of observers. to-sensor links (and to the pairwise Please contact:
keys) would be hard, due to the number Mirosław Kutyłowski
Moreover, we propose a far more of possibilities growing extremely fast Wrocław University of Technology,
sophisticated design in which on each as the number of links increases. Poland
level of routing there are many potential Tel: +48 71 320 21 09
sensors to play the roles of Pi and Ri. In The architecture described here is cur- E-mail:
this case the adversary usually has to rently being analysed from the point of

Ubiquitous Machine-to-Machine
Service Networks
by Johanna Kallio and Juhani Latvakoski

In the near future, there will be many more embedded devices than there are mobile phones. When
these devices are connected to the Internet, many novel kinds of ubiquitous service will be enabled.

It has been estimated that in 2010, the devices and their enabled applications specify a universally applicable M2M
number of communicating devices will in wired and wireless systems, regard- concept that will enable the interopera-
be a thousand times greater than the less of the supplier. Information tech- tion of sophisticated M2M applications
number of mobile phones, which is nology applications usually operate as through heterogeneous wired and wire-
already more than one billion. When separate M2M solutions that are less IP communication networks.
connecting devices such as various unaware of each other. As a result, a
machines, actuators and sensors to the number of business opportunities Made up of seventeen partners, the
Internet, novel types of service are remain unexploited as the services pro- international Usenet consortium
enabled. Previously, such devices com- vided by the devices cannot be placed focuses on M2M research enabling
municated with services using technol- on the Internet. ubiquitous M2M service networks. The
ogy such as SMS. The applications were project consortium is led by the VTT
vendor or domain-specific closed sys- The three-year Usenet project funded and includes industrial, SME and
tems, for which achieving interoperabil- by the Eureka/ITEA2 programme is research partners from Finland, Bel-
ity with other vendor/domain systems developing a service concept for solv- gium, France and Spain.
was challenging. The Usenet (Ubiqui- ing the above interoperability problems.
tous M2M Service Networks) project M2M services refer to the services M2M Architecture
aims to enable ubiquitous machine-to- resulting from collection, transmission The system components provided by
machine (M2M) service networks, in and processing of information, and different suppliers have a strong influ-
which the M2M infrastructure is able to establish an interactive system with the ence on the structure of M2M systems.
connect and combine services produced remote devices that are ultimately inte- M2M systems usually require the inte-
in different domains in an interoperable grated within a managed M2M software gration of components coming from
way (see Figure 1). system. The project has generated new various stakeholders in the value chain:
types of M2M service scenario, which M2M service providers, M2M opera-
The Usenet Project are related to ubiquitous building infra- tors, M2M manufacturers, software
Currently, no universally applicable structure, machine tools, consumer houses and M2M system integrators.
M2M service infrastructure exists that devices, home automation and telemat- The referred components need to be
would allow interoperation between ics domains. The primary goal is to interoperable in order to establish sensi-

30 ERCIM NEWS 76 January 2009

Figure 1:
Ubiquitous M2M service networks.

ble business operations. Traditionally, several communication infrastructures water or electricity, and the space is
M2M solutions have applied vertical for collecting data and controlling monitored with a video camera. The
architecture and closed solutions. This M2M devices. Smart services can be user is able to follow what is happening
has created challenges in the distribu- based on information, which is col- in the system via the home user inter-
tion of added value, which has been a lected from several service platforms. face (Figure 2), and can control the
barrier to M2M market enlargement. It The challenges of horizontal systems space with sensors and actuators con-
has been estimated that horizontal are related; e.g. the overall quality of nected to the space. For example, users
architecture has better possibilities to end user services and security, which can track the outside and inside temper-
boost M2M market. inevitably requires the existence of ver- atures and are provided with alarms and
tical interfaces. warnings of water leaks, unexpected
It is expected that horizontal architec- weather or the presence of a house-
ture will make it easier for different Home Surveillance – an M2M breaker. They also have control over the
players to be part of the M2M value Application Example lights, heating and so on. Automatic
network. For example, an M2M asset A Usenet M2M application scenario has services related to control operations
devices manufacturer can offer control been implemented to demonstrate the can also be included.
and administration services for their Private Space M2M system for residen-
products. Communication infrastruc- tial homes. The purpose of the system is Potential of M2M Systems
ture can be connected to these devices to provide smart surveillance services M2M systems will provide essential
by means of various telecommunica- of private space. Various sensors meas- business possibilities and advantages
tions' manufacturers and service ure quantities such as humidity, temper- for companies, especially when infor-
providers. Service platforms can utilize ature, light levels and consumption of mation systems controlling their core
processes are utilizing the real-time
information produced by an M2M sys-
tem. In consequence, a company can
increase the quality of its services,
reduce costs and increase customer sat-
isfaction. This fundamental change,
which will bring new business opportu-
nities for companies, can already be
seen in the market. VTT aims to help
companies to take advantage of this
rapidly growing M2M market.

Usenet project:

Please contact:
Juhani Latvakoski, Johanna Kallio
VTT, Finland
Tel: +358 40 520 0149
Figure 2: User interface of Usenet experimental M2M application.

ERCIM NEWS 76 January 2009 31

Special Theme: The Sensor Web

Short-Range Communication in Ubiquitous

Professional and Consumer Applications
by Mikko Sallinen, Esko Strömmer and Pirkka Tukeva

Short-range communication technology is proving its worth in many areas of application. Here we
illustrate three case studies involving NFC (Near-Field Communication) and Nanonet radio
technology. These include a consumer application with the TouchMe paradigm, and professional
applications for force measurements on train wheels and acceleration measurements on the axis
of a paper mill machine. We also show the advantages of these short-range communication
technologies and discuss other potential applications.

In recent years, a rapidly growing range higher than in consumer applications. more data-intensive sensor applica-
of wireless communication technology Current applications include measure- tions, NFC can easily establish a Blue-
has become available and is being ment and wireless data transfer from tooth connection between two devices
applied to numerous application the field to a central unit, as well as when they are brought close together. It
domains. One of the latest examples of control and monitoring applications. In is not necessary for a user to find the
short-range communication technology many cases this works well, but the correct menu items and configuration
is a light and simplified version of Blue- problem is often that even if the data parameters on the mobile handset,
tooth known as Bluetooth ULP (Ultra- can be transferred, it is still necessary which makes interaction with the NFC-
Low Power), earlier known as Wibree. to have power cables. Energy harvest- enabled sensor very elegant. On the
Other examples of short-range commu- ing and power management is one of other hand, the sensor must be within
the reach of the user.

For easy piloting of NFC applications

we have developed Smart NFC Inter-
Mobile phone
with NFC and
face, a building block that provides NFC
optional and Bluetooth communication capabil-
Bluetooth ity to various smart objects. Thus it also
NFC- Int erfac er
NFC makes these smart objects capable of
(+BT) communicating with mobile handsets as
depicted in Figure 1. Smart NFC Inter-
face incorporates components for short-
range wireless communication with
mobile handsets, a set of wired inter-
(+BT) faces for sensors and other smart
objects, local data processing and stor-
age, and a clock for time stamping the
measurement results and other events.
NFC- Int erfac er


Smart NFC Interface


The advantages of NFC over alternative

wireless short-range communication
Figure 1: Intelligent interfacing module for smart objects. technologies in sensor applications,
such as Bluetooth, are its lower power
consumption, shorter communication
nication technology include RFID, the key issues in the more challenging set-up latency and better immunity to
Bluetooth, IrDA and Zigbee. From these industrial applications. eavesdropping. NFC can also operate in
technologies, IrDA and Bluetooth offer passive mode with extremely low
a huge number of applications. In this article, we use two different power consumption, which means wire-
technologies. The first is NFC (Near- less sensors can have multiple-year bat-
New technology will also generate a new Field Communication), an RFID-based tery lifetimes. Even batteryless sensors
business for the service sector which is technology that provides short-range taking their power from the RF field
an important growing market. There are communication at 13.56 MHz licence- that is generated by a nearby active
many areas in which wireless technology free band. The second is the Nanonet NFC device are feasible.
will find ubiquitous application, includ- radio module, which uses the 2.4 GHz
ing entertainment, healthcare, automo- licence-free band and is manufactured In addition, NFC enables simple com-
tive and logistics. These are areas which by Nanotron Ltd. munication based on the TouchMe par-
have a huge future market potential. adigm that from the user's viewpoint is
Touch-based Sensor Readout by NFC fast and convenient compared to plug-
In industrial or other professional appli- NFC can be used to read data from a ging in cables or manually establishing
cations, the requirements for accuracy, sensor to a mobile handset by touching a Bluetooth connection between the
reliability and timing are typically the sensor to the mobile handset. In sensor and the mobile handset.

32 ERCIM NEWS 76 January 2009

Figure 2. Force measurements
from a train wheel.

Figure 3:
Measurement system
inside the paper roll.

Force Measurements on Train Wheels presence of surfaces and walls. Commu- with the axis of a paper roll. The goal
A typical challenge in the industry is nication with sensors is carried out using was to measure the acceleration of the
taking measurements on rotating an SPI (serial peripheral interface) inter- axis and be able to predict cases in
machines. The difficulty of wiring both face with a 16 Mhz line. This wireless which the paper will tear in the paper
data and power supply makes this a link is carried out using NanoNET machine. It is known that tear will occur
non-trivial problem. Examples of this power modules by Nanotron. when there is vibration in the axis.
kind of object are wheels, turbines, For this purpose, a specific measure-
blades and paper rolls. In this field, the Nanotron uses 'chirp' modulation tech- ment tool was designed (see Figure 3).
goals are to optimize construction and nology, which provides benefits such as It includes a triple-axis acceleration
maintenance by condition monitoring, resistance to multipath fading. In the sensor and a transmitter to send online
and to be able to foresee the need for receiver side of the measurement sys- measurement signals. The challenge in
maintenance. It would be much more tem, the System-on-Chip platform the task was to successfully transmit a
effective to carry out maintenance receives the data. In each wheel there signal through a full paper roll. For the
based on the actual wear of parts, rather are twenty pairs of strain gauges that data transmission, we used the same
than based on age as tends to be the measure the deformation of the wheel. radio technology as in the train wheel
case. In the future, this will lead to sig- These strain gauges are read at a fre- case; with this we were able to measure
nificant savings. quency of 3 kHz. The measurements are and recognize peaks in the vibrations of
filtered to a frequency of 500 Hz and the axis and the data was transmitted
To take force measurements on train the data is transferred using 48 channels successfully even through the thickness
wheels, strain gauges are placed on the for transformation. It is therefore neces- of the roll.
surface of the wheel in the form of a star. sary to run several radios in parallel and
The measurements are converted using to carry out high-speed synchroniza-
FPGA platform and then transferred tion. The sampling rate is 290 MHz in
using the 2.4 GHz ISM band and 2Mb/s the developed system. Please contact:
wireless link. The power consumption Mikko Sallinen
while in use is 60 mA, and 1μA while in Acceleration Measurements from the VTT Technical Research Centre,
standby. Maximum transit power is 10 Axis of a Paper Mill Finland
mW. The outdoor range of the radio is In the third application, we integrated a Tel: +358 40 7235263
100m; when indoors it depends on the 3D acceleration measurement system E-mail:

ERCIM NEWS 76 January 2009 33

Special Theme: The Sensor Web

The CLARITY Ubiquitous Robotic Testbed

by Gregory O'Hare, Mauro Dragone and Jennifer Treanor

The CLARITY Centre for Sensor Web Technology in Ireland is currently constructing a ubiquitous
robotics testbed by integrating a collective of mobile robots with a wireless sensor network and a
number of portable devices. The new, mixed testbed will be hosted at the School of Computer
Science and Informatics at University College Dublin, (UCD), and will also avail itself of the
laboratory facilities hosted in Dublin City University (DCU) and Tyndall, Cork. The testbed will provide
a service for all researchers interested in developing ubiquitous robot applications.

The number and variety of applications posed advances in mobile networking, and software components are not stable
of robots in our daily environment is on routing, data collection and data analy- but may change at run-time. In order to
the increase. Examples include robotic sis within the WSN community. Robots adapt to such environments, these
pets (eg Sony Aibo), household appli- are just another class of user of sensor applications must exhibit run-time flex-
ances (eg iRobot vacuum cleaner data, potentially harvested both by their ibility, such as the ability to reorganize
Roomba) and assistive technology (eg on-board sensors and by the sensors the interaction patterns of their architec-
the MANUS wheelchair). At the end of already available in their environment. tural elements during execution. In par-
2006, figures for service robots for per-
sonal or domestic use stood at nearly 3.5
million, with this projected to more than
double before the end of 2010.

At this point, a principal challenge for the

robotics community is the integration of
robots into today's digital society. Inex-
pensive Internet access and the diffusion
of wireless computing devices have
made ubiquitous/pervasive computing a
viable reality that augments the normal
physical environment and supports the
delivery of services anytime, anywhere
to human users. Endowing these ubiqui-
tous devices with intelligent behaviour,
and thus creating intelligent environ-
ments, is termed ambient intelligence. Figure 1: Robots at the CLARITY Centre for Sensor Web Technology.

Robots are a compelling instance of

those artefacts that comprise and deliver
the ambient space. Modern multi-robot Robots can also actively help the sensor ticular, Component-Based Software
applications have moved away from the network, not only by acting on the envi- Engineering (CBSE) and Agent-Ori-
historical view of monolithic control ronment, but also by helping to deploy, ented Software Engineering (AOSE)
systems: they run on specific computa- program and maintain sensors; collabo- paradigms are natural candidates to pro-
tional units and are in charge of their rating with sensors' localization; and by vide modular architectures to integrate
own 'hardwired' hardware. They are acting as mobile gateways in multi-hop and dynamically organize the different
already perceived to belong to a larger, networks. They also push the bound- system functionalities.
open distributed network made up of aries of WSN research by requiring
different sensors and effectors. Ubiqui- interoperable and efficient solutions to Ubiquitous Personal
tous robotics further extends this view data collection and online analysis. and Social Assistant Agents
by explicitly addressing the need for Not only will these robots have to deal
interoperability between robots and Adaptive and Self-Organising with a variety of complicated tasks, but
existing ubiquitous and pervasive infra- Software Architectures they will also be expected to behave in
structures, such as wireless sensor net- The need for adaptive and self-organiz- a socially intelligent and individualized
works (WSNs). ing software architecture for ubiquitous manner in order to meet the diverse
robotics emerges from the very same requirements of each user. However,
CLARITY is particularly interested in requirements as for autonomous opera- reconciling the personalization/social
employing the new testbed in three tions. In traditional networked robot aspect with pervasiveness and ubiquity
overlapping areas of research: systems, component integration is remains a largely unexplored area of
essentially an offline feature. In con- research. On both fronts, user interface
Integration between Robots and WSNs trast, these systems demand a more agents, eg acting as a personal agent
The new testbed represents an ideal open and dynamic approach, as the assistant (PAA) to their user, have
arena in which to develop and test pro- nature and availability of their hardware already been widely adopted as intelli-

34 ERCIM NEWS 76 January 2009

gent, adaptive social interfaces to the Ubiquitous robot systems involve many (i) visualize the topology and behaviour
digital world, eg in the form of virtual interacting hardware and software com- of the WSN through a network map; (ii)
characters interacting with the user via ponents, and malfunctions in both types log collected parameters for network
PCs, PDAs and the Internet. As such, of component can encumber them. analysis; (iii) localize nodes on a pro-
the experience accumulated in these Such characteristics pose a considerable vided 2D floor plan; and (iv) formulate
applicative domains may be used to challenge to the application of a strong and inject composite queries into the
inform robotics research. Moreover, as engineering perspective within robot- network.
both software agents and robots ics, as it is difficult to correctly gauge
increasingly inhabit the same human the accomplishment of specific system Links:
social space, their mutual interaction objectives and guide improvements and CLARITY Centre for Sensor Web
and combined operation within human further developments. For these rea- Technologies:
societies will also acquire increasing sons, our research has commenced by
importance. defining a number of logging and CLARITY Ubiquitous Robotic
inspection facilities. Testbed:
The New Testbed IFR Statistical Department. 2007 world
The new testbed integrates and extends On the robot side, an XML service robotics survey:
some pre-existing facilities, specifically: developed in collaboration with the
• WSN of 70 Berkeley motes measur- Interoperable Systems Group at DCU The Interoperable Systems Group at
ing humidity, light and temperature will provide a generic logging service DCU:
• ten mobile robots, equipped with an for the instrumentation of both CBSE
array of state-of-the-art sensors, and AOSE frameworks. In addition to Octopus:
including USB cameras, laser range offline analysis, the service will also
finders, sonar, infrared, odometers support hardware-in-the-loop simula- dashboard-sensor-networks-visual-
and bumpers. Each robot carries a tions (HILS), thanks to synchronization control
mote able to measure ambient vari- and replay functionalities that will
ables, which is also equipped with allow focusing online analyses and sim- Please contact:
triple-axis accelerometers, magne- ulation over isolated groups of software Gregory M. P. O'Hare
tometer, compass and microphone modules. CLARITY Centre for Sensor Web
• a variable number of Internet gateways Technologies, University College
• a variable number of PDAs and mobile The Octopus interactive dashboard, Dublin, Ireland
phones equipped with Bluetooth. developed at UCD, will be employed to E-mail:

Beyond RFID:
The Ubiquitous Near-Field Distributed Memory
by Paul Couderc and Michel Banâtre

The upcoming radio frequency indentification (RFID) revolution will undoubtedly contribute to the
blending of the information society with the physical world: while common 'communicating objects'
are currently restricted to complex electronic devices such as cell phones, cameras etc, RFID is in
fact able to promote anything as a communicating object. In such a new world, data will not
necessarily always flow into computers and networks, but may be physically retained by objects
moving in real space.

While RFID technology promises many concept of global identification, associ- For example, security procedures in air-
useful applications, such as improved ated directory services or tracking data- ports require that your personal effects
safety, easier and faster interactions, bases. However, technically RFID are are checked separately from you by X-
reduced error in data input and automa- just small memory devices that can be rays. Forgetting one of your items, or
tion of tedious processes, it also raises addressed by near-field communica- mistakenly exchanging a similar item
serious concerns – in particular the pri- tion, and although identification has with someone else occurs frequently.
vacy issue. In a world where many per- been the main application target, these Solutions have been proposed for this
sonal objects are electronically identi- devices provide support for alternative problem, based on active tags attached
fied, the activities of individuals could mechanisms. to the items that are monitored by an
be traceable in a similar, though much owner tag. This is impractical for sev-
more comprehensive, way to 'googling' We will now present such an alterna- eral reasons: active tags are expensive,
someone today on the Internet. tive, through the example of the 'Ubi- they require batteries (and hence regu-
Check' service, a solution to a common lar maintenance), radio emissions may
An important cause of this issue is that problem when travelling: it is unfortu- be restricted by regulations (on planes
RFID systems are usually based on the nately quite easy to forget something. for example), and temporarily separat-

ERCIM NEWS 76 January 2009 35

Special Theme: The Sensor Web

ing an item from its owner would the privacy concerns of many other Another application example is to use
require the alarm to be disabled. RFID approaches. Specifically, tracking RFID to build distributed maps for
of individuals is not easy, since the tags' robotic support. In the Roboswarm
Ubi-Check represents another solution content may change often for the same project we consider very simple robots
using RFID tags attached to the items. It person and same set of objects. Further, with limited knowledge and sensors. In
is possible to write in the memory of the the system is not based on identifica- order for them to navigate, RFID tags
tags the data required to check the tion, ensuring greater privacy. are arranged in the space where the
integrity of the group of items. One pos- robots are working. Each tag contains
sible implementation is to compute a Another interesting aspect is that check- relative spatial references pointing to
digital certificate from the identifiers of points and association points are the nearest tags, effectively making a
all the items. An important aspect is that autonomous and only carry local pro- graph. These pointers allow a robot to
the identifiers associated with each item cessing. The system is therefore not reach one tag from another, assuming
can be regenerated regularly (eg for that odometry drift is low enough while
each trip): they are only used for a moving the distance separating two
locally computed integrity check, not tags. In this distributed map example,
for identification. The values could be the memory is distributed quite thinly
written in the tags at, for example, the over the physical space. As near-field
airport check-in, the train station, or communication technology gets better
even when leaving home. Then, at rele- and cheaper, denser and more ubiqui-
vant points after the area in which peo- tous memories are likely to emerge in
ple are vulnerable to item loss or the environment, with computing appli-
exchange, we deploy checking gates cations that we are yet to imagine.
(such as the exit of the security check in While the industrial promoters of
airports, or the exit gate of a plane or a RFID/NFC are mostly concentrating on
train). These gates would ensure the applications based on identification,
integrity of groups of items crossing almost anything can be done with mem-
them, warning people in the case of a Figure 1: Ubi-Check: integrity checking ory devices, including systems that
missing item or the presence of some- using a collection of eletronic tags respect privacy. A key question to mak-
one else's item. ing this possible is: will there exist
completely 'blank' or free-format tags,
This solution is a distributed system dependent on a remote information sys- without any pre-existing and non-
where only local properties are checked tem. This has important benefits in rewritable identifier?
in order to ensure a global goal. In fact, terms of extensibility, reliability and
it uses a principle similar to the trans- deployment costs. Link:
mission of a file in independent frag-
ments over a packet network, where This service is an example of a more
integrity is verified by checking general usage of near-field communica- Please contact:
sequence number coherency or check- tion beyond identification. As men- Paul Couderc
sums, except that here the data are car- tioned previously, this technology con- Centre de recherche INRIA Rennes -
ried by 'physical' fragments. Such a sists of memory devices, a basic and Bretagne Atlantique, France
solution is interesting because while generic support for computing, obvi- Tel: +33 2 99 84 72 92
providing a security service, it avoids ously not limited to storing identifiers. E-mail:

Rapid Prototyping of Sensor-Based Applications

with SunSPOTs
by Manfred Bortenschlager, Elisabeth Haid and Andreas Wagner

Sensor technology has the potential to boost productivity just as the Internet did. We demonstrate
the opportunities presented by sensors and sensor network technology by deploying a corresponding
framework in an indoor environmental quality application. Our framework is based on OGC Sensor
Web standards and exploits SunSPOT sensor technology for rapid prototyping.

Due to the restrictions of industrial sen- bility to different domains. As a result, and for testing and verifying algorithms
sors, the engineering of applications that testing, extending and porting an appli- on a small scale prior to deploying them
exploit sensor network technology is cation based on a sensor network is in industrial operation. We demonstrate
difficult. Such sensors are usually pro- expensive. Java-based SunSPOT sensor this methodology in the development of
prietary and inflexible as regards pro- technology has been developed for an indoor environment quality (IEQ)
grammability, reusability and applica- rapid prototyping of such applications assessment application that exploits

36 ERCIM NEWS 76 January 2009

SunSPOTs. This application is based on This tool can be exploited to reduce tration of the sensors, and reading and
a wireless sensor network, adopts OGC costs by evaluating novel algorithms a processing of sensor data; for example,
Sensor Web standards, and provides a priori before adapting them to real- users can visualize and interaction with
visual interface to the sensor values. world problems. current sensor values on a graphical,
Web-based interface. In particular, the
SunSPOT Sensor Technology This form of sensor technology will Sensor Model Language (SensorML),
SunSPOTs are embedded hardware prove useful in a broad spectrum of Observation & Measurements (O&M),
modules that are equipped with a 180 applications and domains, such as early and Sensor Observations Service (SOS)
Mhz CPU, 512 KB RAM, 4MB flash detection and warning systems, envi- specifications are adopted.
memory, three on-board sensors (tem- ronmental monitoring, automotive
perature, light and three-axis accelerom- engineering, warehouse/container man- Communication between the sensor and
eter), hardware interfaces for the inte- agement, logistics, monitoring of build- the base station can occur in both push
and pull modes and in a regular or on-
demand fashion, where the values are
communicated over the meshed wire-
less sensor network. All configurations
can be defined by the user during run-
time. In addition, a user can employ a
sensor value reader device (essentially
another SunSPOT) in order to get data
from a specific sensor by physically
moving into the communication range
of that sensor and querying the accor-
dant data.

Future Activities
Further research will focus on the
enrichment of the visualization compo-
nent by correlating it with information
SunSPOT sensor. potentially coming from other, external
sources such as the Internet. This
should be achievable with reasonable
gration of arbitrary external sensors, and ings, home automation, weather fore- effort due to the adoption of open Sen-
the IEEE 802.15.4 wireless transmission casting, medical monitoring of patients sor Web standards. In addition, method-
technology, which is enabled for mesh and diagnosis, and agriculture and ologies will be investigated that will
networking. SunSPOTs are entirely pro- farming. allow algorithms and functionalities
grammable in the Java programming that were successfully verified on a
language and thus help to abstract from Application: Indoor Environmental SunSPOT platform to be ported to non-
the underlying hardware. No direct Quality Measurement Java sensor platforms with minimal
interaction by machine code with the Based on SunSPOT sensor technology effort.
hardware is necessary, which signifi- we developed an indoor environmental
cantly eases the development of sensor- quality application. Factors that influ- The work presented here is part of the
based applications. ence the climate in a room and con- MobilityLab, a Centre of Excellence on
tribute to human well-being are meas- the Engineering of Location-Based Sys-
The SunSPOT sensor technology was ured at different positions in a room. tems, and involves Sun Microsystems,
developed primarily to satisfy three tar- The sensor data, then, can be retrieved Vienna University of Technology (Insti-
get groups: education (introducing from the sensors and are further tute for Geoinformation and Cartogra-
pupils and students to related topics processed in our sensor network. phy) and Salzburg Research.
such as programming, networking,
embedded systems, robotics or hard- Each sensor station is composed of two
ware), research and development main components, namely (i) an exter- Links:
(allowing for fast implementation of nal sensor which can measure electro-
systems that integrate sensor technology magnetic pollution, air pressure, humid-
and for easy testing of their behaviour), ity, air temperature, brightness, noise or /groups/sensorweb
and hobbyists (developing small sensor- carbon dioxide, and (ii) a SunSPOT
based applications for personal needs). module which is responsible for pre-
processing acquired sensor data and Please contact:
One of the major design goals of propagating them through the sensor Manfred Bortenschlager
SunSPOTs was to provide a tool for network. Salzburg Research
rapidly prototyping sensor-based appli- Forschungsgesellschaft mbH / AARIT,
cations, and for testing and verifying The base station managing this sensor Austria
algorithms on a small scale prior to network is an OGC-compliant Sensor E-mail: manfred.bortenschlager@
deploying them in industrial operation. Web application. It allows for adminis-

ERCIM NEWS 76 January 2009 37

Special Theme: The Sensor Web

Building an Adaptive Environmental Monitoring

System Using Sensor Web Technology
by Jer Hayes, Greg O'Hare, Harry Kolar and Dermot Diamond

The ultimate goal of environmental sensor networks is to realize the concept of an 'adaptive environment'
– one that senses and rapidly adapts to potential incidents in order to minimize their impact.

Environmental pollution affects human

health and reduces the quality of our
land and water. As a result, there is great
interest in monitoring water and air
quality and ensuring that all areas are
compliant with legislation. Ubiquitous
environmental monitoring places con-
siderable demands upon existing sens-
ing technology. The combined chal-
lenges of system longevity, autonomous
operation, robustness, large-scale sensor
networks, operationally difficult deploy- Figure 1: The phosphate analyser developed by CLARITY (left) and results over an 8-day period
ments and unpredictable and lossy envi- from a field trial at a waste water treatment plant (right) showing excellent agreement with a
ronments collectively represents a tech- reference analytical system (total duration 49 days).
nological barrier that has yet to be over-
come. The CLARITY Centre for Sensor
Web Technologies is working with IBM
to confront these challenges. Ubiquitous critical for both health reasons and for pled with significant reductions in their
sensing envisages many aspects of our the sustainable exploitation of these cost is the key to enabling scaled-up
environment being routinely sensed. resources. Sensor Web technology such deployments of sensors at multiple
This will result in data streams from a as low-power wireless communica- locations, and this is a key goal for the
large variety of heterogeneous sources, tions, coupled with the emergence of CLARITY-IBM team.
which will often vary in their volume new and reliable sensors such as the
and accuracy. The challenge is to microfluidic analyser platform devel- Sensor networks provide a web of inter-
develop a networked sensing infrastruc- oped by CLARITY (see Figure 1), now connectivity: multiple sources of infor-
ture that can support the effective cap- enable environmental data to be col- mation that will allow decision-making
ture, filtering, aggregation and analysis lected at much higher temporal and spa- processes to be more accurate and effi-
of such data. This will ultimately enable tial resolutions. Other examples of sen- cient. These processes can be complex
us to dynamically monitor and track the sors being developed by CLARITY and demanding however, and are often
quality of our environment at multiple researchers are very low-power and constrained in a number of possibly
locations. The ability to monitor quality low-cost colorimetric sensors that can conflicting dimensions such as quality,
is a prerequisite to maintaining quality, be fabricated by modifying commercial responsiveness and cost. CLARITY
and ensures that detected pollution inci- optical components with chromo- researchers and IBM are working
dents are dealt with as quickly as possi- responsive films. For example, inex- together to examine in-situ decision
ble, dramatically minimizing their pensive yet sensitive chemo-sensors making, whereby decisions are effected
impact. In effect, the ultimate goal of can be made by applying such coatings based upon inferences made from both
environmental sensor networks is to to light emitting diodes (LEDs), as locally sensed data and data aggregated
realize the concept of an 'adaptive envi- shown in Figure 2. Detailed experimen- from sensor networks. The in-situ sen-
ronment' – one that senses and rapidly tation with these devices in the field has sor nodes that comprise a sensor net-
adapts to potential incidents to minimize demonstrated their effectiveness at work are often computationally chal-
their impact. detecting and tracking the dispersion of lenged with respect to processing
volatile chemical plumes. Continued power, as they are developed to be low-
In recent years an increasing number of improvement in the long-term reliabil- cost, low-power devices rather than
environmental incidents have occurred ity of chemical sensing platforms, cou- high-performance computing devices.
which would have benefited from an
adaptive environment approach. These
include the detection of lead in water
supplies, cryptosporidium outbreaks Figure 2: A gas sensor made by
and, as recently reported by the Irish coating a pH indicator dye (left) and
EPA, and high pollution levels in a third the sensor platform, mica2dot mote
of Irish rivers and streams. In Ireland (right). In the presence of an acidic
and the rest of the world, proper man- vapour/gas, the sensor changes
agement of environmental resources is colour from deep blue to yellow.

38 ERCIM NEWS 76 January 2009

This computational challenge places a seaboard of Ireland. Collaboration in the filtering, aggregation and analysis of
major constraint upon the reasoning development of SmartBay will form a environmental data. While the initial
process. To overcome this challenge, core activity of the Global Centre of real-world benefits of such technology
CLARITY and IBM are developing a Excellence for Water Management, will be in the environmental domain,
hybrid reasoning approach to deliver in established by IBM in Ireland in 2008. these benefits will ultimately belong to
situ decision-making that combines SmartBay-related research will mirror any application area of the Sensor Web.
stream-based computing with multi- the monitoring activities in the Hudson
agent system techniques. The hybrid River in New York State of the Beacon Links:
reasoning approach builds on System S, Institute for Rivers and Estuaries, and CLARITY:
an IBM technology for distributed will in part compliment the River and
stream processing, and Agent Factory, Estuary Observation Network (REON) IBM Big Green Innovations:
developed by CLARITY researchers to established for that purpose.
provide a powerful collaborative deci- greeninnovations/
sion-making system. In partnership The application of Sensor Web technol- SmartBay:
with the Irish Marine Institute, this ogy to environmental sensing will even-
hybrid approach is being tested through tually result in the realisation of the operational/SmartBay/SmartBay.htm
an environmental demonstrator project 'adaptive environment' concept, though
entitled SmartBay, which seeks to networks that can rapidly detect and Please contact:
deliver an infrastructure to validate dis- adapt to potential environmental inci- Dermot Diamond
tributed in-situ real-time environmental dents, in order to minimize their impact. CLARITY Centre for Sensor Web
monitoring technolgies. CLARITY and IBM are demonstrating Technologies, Dublin City University,
key technological building blocks of the Ireland
SmartBay is a next-generation water adaptive environment through projects E-mail:
management system (both marine and such as SmartBay, through which new
freshwater), aspects of which are cur- sensing and collaborative decision- Harry Kolar
rently under development by CLARITY making technology is being used to test IBM, T.J. Watson Research Center,
for the Marine Institute of Ireland. It is elements of a networked sensing infra- USA
located at Galway Bay on the Atlantic structure that can support the capture, E-mail:

Remote Water Monitoring With Sensor

Networking Technology
by Thiemo Voigt, Nicolas Tsiftes and Zhitao He

Sensor networks enable remote monitoring of natural environments such as glaciers, volcanoes and
bodies of water. Within the project 'Sensor Networks to Monitor Marine Environment with Particular
Focus on Climate Changes', SICS and partners are designing and implementing flexible,
reprogrammable sensor network solutions suitable for monitoring the marine environment with high
resolution in time and space.

Carrying out marine research requires works to Monitor Marine Environment of material on hard surfaces in aquatic
that studies be undertaken in remote with Particular Focus on Climate environments. We first intended to use
environments such as the Baltic Sea. Changes', SICS and partners have an oil-filled garage, but found that this
However, marine environmental moni- designed an advanced water monitoring affected the sensor readings. Initial
toring is expensive: the cost of operation system (Figure 1). Our system features experiments with an air-filled garage
at sea includes at least €10 000-15 000 a diving unit, consisting of a bin con- have shown that this solution also pre-
per day for the use of a lar ge research taining several sensors connected to one vents fouling.
vessel, plus the cost of laboratory tech- sensor node. The diving unit moves up
nicians, analytical instrumentation and and down an anchor line. In this way we Our system is driven by rechargeable
logistics. As a result, the marine envi- are able to take measurements at any batteries. Previous projects have had
ronment is poorly monitored. For exam- depth using only one set of sensors. A problems with solar power. Our project
ple, in the Baltic Sea's Bothnian Bay, the pressure sensor determines the depth at partner Uppsala University therefore
nine fixed stations are visited only eight which measurements are being designed a wave energy generator to
times a year. recorded. When the diving unit is not in obtain electric power from vertical
the water taking measurements, it is wave movements. As a ring of stacked
Sensor networks mean that marine data parked in a garage that is part of our magnets floats up and down along a rod
collection can be undertaken in a much buoy construction. By parking the div- wound with metal coils, an alternating
more cost-efficient fashion. Within the ing unit in the garage, we expect to voltage is induced across the two ends
multi-disciplinary project 'Sensor Net- eliminate fouling, ie the accumulation of the coil. This is further rectified and

ERCIM NEWS 76 January 2009 39

Special Theme: The Sensor Web

cost of the transferred data, we com-

press the data before sending it. Previ-
ous work has shown that compression is
much cheaper than communication for
typical sensor nodes with low-power
radios. The same is true for the more
power-hungry GPRS communication.
To this end, we have designed a new
compression algorithm called SB-ZIP
that is more efficient than state-of-the-
art compression algorithms. For exam-
ple, SB-ZIP compresses 4.5 MB of
acceleration data measured on the buoy
Illustration: Bo Reinerdahl

to only 1.5 MB, whereas the well-

known S-LZW algorithm is only able to
reduce it to slightly over 3 MB.

Figure 1: Sensor system for remote water monitoring. SICS is running this project in collabo-
ration with Umeå Marine Sciences Cen-
tre, Uppsala University and SMHI, the
filtered to provide a constant charging energy. The sheer amount of data Swedish Meteorological and Hydrologi-
voltage to the batteries. exceeds the limited RAM available in cal Institute. At the conclusion of the
sensor nodes. Moreover, to handle the project in late 2009, we expect to have
Our system is designed to cope with different data types smoothly, we use deployed around ten buoys at places rel-
water depths of around 100m, meaning Contiki's Coffee file system to store evant for marine monitoring in the
the number of measurements taken dur- data. The data is transferred onshore Baltic Sea.
ing one dive can be significant. Storage using GPRS (General Packet Radio Ser-
is thus required for the large volumes of vice), and this is too energy-consuming
measurement data: not only those col- a task to be performed after every dive. Link:
lected by the diving unit but also from We have therefore ported Coffee to
light and acceleration sensors placed on operate on SD cards that enable cheap
the top of the buoy. System data also mass storage. Please contact:
needs to be saved; for example, the Thiemo Voigt, SICS, Sweden
energy produced by the wave energy Due to the high energy consumption of Tel: +46 8 633 1598
generator as well as the available the GPRS unit as well as the per-byte E-mail:

Sensors Anywhere – Sensor Web Enablement

in Risk Management Applications
by Gerald Schimak and Denis Havlik

The increasing frequency, severity and consequences in Europe of floods, storms, forest fires and
other natural hazards sensitive to climate change has clearly shown the shortcomings of existing
environmental monitoring and information systems. The observed inefficiency is primarily a
consequence of historical and organizational factors. An exorbitant amount of work on data and
service standardization would be required to build more efficient information systems using state-
of-the-art technology.

Emerging technology in risk monitoring the currently available and emerging Sensors Anywhere (SANY) is an ambi-
and management has the potential to technologies offers rapid deployment, tious FP6 IST Integrated Project deal-
speed up the necessary organizational easy maintenance, quality assurance and ing with sensor networks for environ-
and structural changes. Ad hoc wireless automated data processing along the mental and risk management applica-
sensor networks and the collective intel- whole information processing chain tions (Figure 2). SANY aims to con-
ligence of the Sensor Web; the plug-and- from smart sensors and wireless ad hoc tribute to joint efforts of the European
measure paradigm of IEEE 1451 smart sensor networks, over automated data Commission (EC) and the European
sensors; the Semantic Web; and the OGC loggers and value-added middleware Space Agency (ESA) on 'Global Moni-
Sensor Web Enablement architecture: all services, to user applications capable of toring for Environment and Security'
these address critical factors of the state- dynamically integrating all available (GMES) by improving the interoper-
of-the-art technology. However, none of data sources at run time. ability of in situ sensors and sensor net-

40 ERCIM NEWS 76 January 2009

The first validation subproject (Fig-
ure 1) illustrates the use of SensorSA
services at data acquisition level, as
well as vendor independence and the
feasibility of building virtual networks
across administrative and technical bor-
ders. Furthermore, it is concentrating on
assessing air pollution episodes, track-
ing pollution back to its source and pre-
dicting the air pollution in urban areas
or around industrial zones.

The second validation subproject con-

centrates on coastal water management
issues, such as assessing, modelling and
predicting bathing water quality with
user-determined 'what-if' scenarios. It
uses advanced data fusion services to
demonstrate the feasibility of run-time
binding of heterogeneous data sources
and the run-time choice of data fusion
service, and to predict microbiological
contamination through data fusion
using all available real-time measure-
Figure 1: SANY air quality management subproject. ments and the results of microbiological
water sample analyses. The real-time
works and taking up the challenges 'observation' data and meta-information, data sources used by this subproject
mentioned above. This means specifi- the so-called 'Sensor Observation Ser- include meteorological data, sea tem-
cally that the Sensor Service Architec- vice' (SOS); a service for sensor plan- perature, turbidity, salinity, chlorophyll
ture (SensorSA) developed in SANY ning and executing tasks, called the 'Sen- and dissolved oxygen levels.
shall allow quick and cost-efficient sor Planning Service' (SPS); a service
reuse of data and services from cur- that allows users to subscribe to specific Finally, the third validation subproject
rently incompatible sources in future alert types, known as the 'Sensor Alert concentrates on geohazard monitoring
environmental risk management appli- Service' (SAS); and a service that facili- in a complex urban environment, which
cations across organizational, adminis- tates asynchronous message interchange means assessing the structural instabil-
trative or regional borders. between users and services, and between ity of architectural objects caused by
two OGC-SWE services, called the 'Web human activities (eg tunnel excavation).
In order to assure the sustainability of Notification Service' (WNS). It demonstrates the integration of wire-
the project results, SANY reuses the less ad hoc sensors into SensorSA net-
open standards from W3C, OASIS, ISO Three validation subprojects of SANY works; rapid deployment, auto-configu-
and Open Geospatial consortium demonstrate the feasibility of building ration and self-management; and data
(OGC). One of the most promising risk management applications with fusion of in situ and Earth Observation
standardization efforts is currently hap- SensorSA. data sources.
pening within the Open Geospatial
Consortium Sensor Web Enablement The authors wish to thank the Informa-
initiative (OGC-SWE). The goal of tion Society and Media Directorate
OGC-SWE is to enable all types of Web General of the European Commission
and/or Internet-compatible sensors, (DG-INFSO) for co-funding the inte-
instruments, and imaging devices to be grated project SANY within the area of
accessible and, where applicable, con- ICT for Environmental Risk Manage-
trollable via the Web. The OGC-SWE ment.
vision is to define and approve the stan-
dards foundation for interoperable
Web-based sensor networks. For this Link:
purpose, OGC-SWE specifications
include a standardized model for repre-
senting and exchanging observation Please contacts:
results as well as an information model Denis Havlik
and encodings that enable the discovery Project co-ordinator S@NY
and tasking of Web-resident sensors. Austrian Research Centers – ARC
Of special interest for SANY are the fol- Figure 2: 'SANY'-trees - sensors anywhere in Tel: +43 50550 3157
lowing: a service for retrieving sensor our environment. E-mail:

ERCIM NEWS 76 January 2009 41

Special Theme: The Sensor Web

FLASH: Fine-Grained Localization in Wireless

Sensor Networks using Acoustic Sound
and High-Precision Clock Synchronization
by Evangelos Mangas and Angelos Bilas

Sensor localization is an important component and enabler of many applications using intelligent
sensors. FLASH is a system that achieves fine-grained localization using acoustic sounds and high-
precision clock synchronization via radio frequency (RF) communication. Sensors can dynamically
localize themselves in space by maintaining synchronized clocks and measuring time of arrival for
acoustic sound pulses. Our techniques focus on achieving highly accurate synchronization and
consistent sound detection. Furthermore, FLASH does not require external infrastructure such as
fixed equipment, specialized hardware support, or great resource consumption. Experimental
results show that FLASH localization is accurate to within 11cm in a variety of indoor environments.

Localization is an important component are sensing the audio frequency spec-

of many sensor applications. For trum. All listening nodes attach a time-
instance, the ability to accurately locate stamp to the received sound pulse, in a
humans or objects in enclosed spaces global synchronized timescale. These
can help in detecting and responding to (synchronized clock) timestamps are
abnormal situations. Techniques such as used to calculate the time of flight for
the use of GPS, image processing or each sound pulse. Distance is then esti-
static ranging mechanisms are either not mated based on the speed of sound. Our
effective, cumbersome to use, or incur work is therefore focused on high-accu-
high costs. racy clock synchronization and sound
detection for range estimation.
Our work focuses on achieving fine- To achieve precise clock synchroniza-
grained localization without excessive or tion we implement a synchronization
specialized resources that use wireless protocol operating at the Media Access
sensor nodes. We rule out the use of Control (MAC)-Layer that does not
extensive external infrastructure, since it introduce significant communication
is usually both expensive to acquire and costs and that deals with fixed over-
complex to install. Moreover, it is impor- heads introduced by the interrupt mech-
tant to achieve localization by using only anism and RF communication. We also
typical resources available in sensor introduce an external mechanism for
nodes, without the use of specialized testing synchronization precision, caus-
hardware peripherals. For these reasons, ing simultaneous interruptions to syn-
we choose an approach that relies on an chronizing motes and then comparing
acoustic sound-ranging scheme. Each the synchronized timestamps produced.
reference node produces an audible Figure 1: Performing experiments with sound We implement our synchronization pro-
sound pulse while the rest of the nodes in our lab. tocol on Mica2dot motes. Our experi-

Figure 2:
Left: Error in estimated distance with respect to absolute distance;
Right: Localization in two dimensions, the sounders are node0 and node1.

42 ERCIM NEWS 76 January 2009

mental results show that synchroniza- timestamps at the beeper node and the compared to methods that use multiple
tion precision has an error of less than listening node results in the time of sensor devices per node.
5μs (median) for a 30s resynchroniza- flight for the sound pulse.
tion period. Future work in the area includes local-
Overall, FLASH demonstrates that ization in outdoor and more demanding
Our acoustic sound technique focuses localization can be achieved with inex- environments, such as in the presence
on consistently detecting the start of the pensive off-the-shelf devices and yet be of obstacles between nodes, intense
arriving sound pulse. Attaching a time- quite precise. Figure 1 shows part of our noise in a room, temperature and
stamp at the beeper node is simple: experimental setup, whereas Figure 2 humidity variations and outdoor envi-
right before applying voltage to the shows our results for 1D and 2D local- ronments. We believe that FLASH and
embedded buzzer we use the local (syn- ization. Using the Berkeley Mica2dot similar techniques will play an impor-
chronized) clock to broadcast an RF motes, the default microphone that tant role in cyber-physical systems and
message. At each listening node we already exists on the sensor platform and in our efforts to better interact with and
identify a sound pulse produced by the cheap simple buzzers, we were able to control our environment.
beeper node by using the periodicity of locate nodes at distances of up to 10m,
the sound pulse and requiring the aver- depending on surrounding noise. The Link:
age peak-to-peak amplitude to surpass a average error in localization precision is
certain predefined dynamic threshold, 11cm for distances up to 7m. However,
which is less susceptible to reflections. our approach does not require either cal- Please contact:
Then we attach a timestamp to the first ibration or any special infrastructure. Angelos Bilas
peak-to-peak measurement that was of Furthermore, our method requires a sin- ICS-FORTH, Greece
greater value than our dynamic thresh- gle sounder and microphone per node, E-mail:
old. The difference between the two resulting in better energy efficiency

High-Density Wireless Geophone Networks

for Oil and Gas Monitoring and Exploration
by Stefano Savazzi, Vittorio Rampa and Umberto Spagnolini

Strong fluctuations in crude oil prices and the expected production peak of current reservoirs are
pushing oil companies to increase their investment in seismic exploration. Replacing cabling with
wireless technology should radically improve the quality of depth imaging and simplify acquisition
logistics. Recent advances in Wireless Sensor Networks (WSN) now allow the wireless community
to satisfy the rigid constraints imposed by seismic acquisition systems, which have a large
number of sensors (> 10 000) over the monitoring area (> 5km2).

Strong fluctuations in crude oil prices oil exploration. Technical limitations in sor/geophone with an error of less than
are pushing oil companies invest more the data-rate efficiency, interference and 1m to avoid degradation of the depth
in seismic exploration of new oil reser- battery use of current short-range wire- imaging quality.
voirs and in new technology to improve less network architectures (eg WiFi,
the quality of depth imaging. Seismic Bluetooth) forced previous proposals Network Architecture
prospecting requires a large number of for wireless geophone system architec- As shown in Figure 1, the proposed
sensors (up to 30 000), such as geo- tures to choose a combination of wire- WGN architecture exploits different
phones or MEMS-based (Micro Electro- less and wired configuration. However, radio transmission technologies to effi-
Mechanical Systems) accelerometers. recent advances in WSN technology ciently handle both short-range trans-
These are deployed over large areas (up conveniently address the issues related missions (ie for short-distance low-
to 30km2) to measure the back-scattered to the strong constraints imposed by power communication among geo-
wavefield generated by an active excita- seismic acquisition systems. A Wireless phones/sensors), and long-range trans-
tion source. A storage/processing unit Geophone Network (WGN) must sup- missions (ie for seismic data delivery to
(sink node) collects measurements from port multiple acquisition settings and storage units and geophone remote
all the geophones in real time to obtain applications. Basic network require- monitoring) that must cover distances
an image of the sub-surface. Current ments are: i) network throughput of of several kilometers. The hierarchical
telemetry is cable-based and usually 150kbps down to 50kbps for single network design requires the deployment
requires hundreds of kilometers of component sensors; ii) real-time (or of a number of Wireless Geophone
cabling, which results in delays, high near real-time) acquisitions with strong Gateways (WGGs) to collect data read-
logistic costs and low imaging quality. delay constraints; iii) remote control by ings from a large number of wireless
sink node and synchronous acquisition geophones (WGs) and forward the data
Wireless technology is thus expected to with a maximum timing skew of 10μs; to the storage unit (SU). These WG
significantly improve the efficiency of and iv) accurate positioning of each sen- nodes are self-organized into independ-

ERCIM NEWS 76 January 2009 43

Special Theme: The Sensor Web

ent sub-networks; ideally the number of OFDM or MultiBand Orthogonal Fre- mission is organized in superframes
devices per sub-network should be as quency Division Multiplexing) provide with the beacon period (BP) carrying
high as 300 nodes to minimize the num- wireless devices with high data rates the essential information of each
ber of WGGs. This results in an aggre- over short ranges of up to 480Mbps, device. Logical device/sensor groups
gated (per sub-network) throughput of and low power consumption (ie below are dynamically formed according to
about 45Mbps (up to 60Mbps). Data 100mW in active transmission mode WiMedia protocol to facilitate the shar-
delivery within one sub-network is but down to 20μW in power-save ing of resources, while wireless
obtained by multi-hop transmissions mode). The MB-OFDM processing can medium reuse can be exploited over
towards the WGGs; WG sensors are also guarantee network scalability different spatial regions.
within 5-100m of inter-node distance to through time and frequency division by
reduce both energy consumption and allowing the use of multiple sub-bands WGG supports specific extended func-
increase battery life. to separate the co-located sub-net- tions compared to a standard WiMedia
works, and coexistence with other 2.4 device. These functions allow: i) the
Physical and MAC Layer Requirements GHz-based radio devices without sig- Gateway to behave as an intermediate
The requirements of self-localization nificant cross-interference. sink, forwarding data to the storage/
and frame synchronization make Ultra processing node SU and controlling
WideBand (UWB) technology the natu- The high number of devices per sub- each sub-network; ii) contention-free
ral choice for short-range transmissions network and the large network size sug- resource negotiations to guarantee real-
within each sub-network. To achieve gest the adoption of a number of distrib- time constraints (eg quality of service
positional accuracy with errors less than uted MAC (Medium Access Control) and maximum delay); and iii) coexis-
1m, the travel-time estimation error for functionalities. Network topology tence of long/short range transmissions.
ToA-based (Time of Arrival) position- should define a hierarchical structure Figure 2 illustrates the MAC layer fram-
ing must be in the order of 3ns with a where the WGG acts as an intermediate ing structure adopted for each sub-net-
minimum required signal bandwidth of sink towards the storage unit. The work, while the probability of full net-
500MHz. UWB technology provides WiMedia standard (ECMA-368 from work coverage versus the BP length is
data acquisition, synchronization and ECMA International, the European shown at the bottom of the same figure.
localization without the use of fully Association for Standardizing Informa- Sensors/geophones are assumed to be
GPS-based (Global Positioning Sys- tion and Communication Systems) has deployed according to the requirements
tem) WGN nodes. Moreover, recent been chosen as the reference for the of a conventional seismic survey. Geo-
advances in radio design (ie MB- development of the WGN MAC. Trans- phone deployment has a major impact

WG #1 Beacon Slot Super Frame

1) Device Address Beacon Period Data period
and spatial position α 1−α
2) Neighbor Information
3) WGG device address
(sub-network ID)
4) Reserved Slots
WGG’s HOBS Unused BSs
3 Slot
WGG 1 2 3 4
Location Time Seismic
Shot line

Reserved Slot
info stamp data
for WG #1
1) Owner Address
WGG 2) Destination Data from WG#2
SU Address
Full coverage probability

Δ x = 20 m
Δ x = 11m Δx =15m Δ y = 6m
Δ y = 6m Δy = 6m ( M = 164 WGs )
( M = 292 WGs ) (M = 216WGs)
WGG: Wireless Geophone Gateway 0.4
WG: Wireless Geophone 0.3
20 m

Δx y
SU: Storage unit 0.2
0.1 WGG
Short range communication α = 12% α = 8% α = 4%
Long range communication Beacon Period length
Figure 1: Wireless Geophone Network architec- Figure 2: MAC layer framing structure (top) used by each sub-network, and its
ture. impact on full network coverage (bottom).

44 ERCIM NEWS 76 January 2009

on the framing structure design: a lower Links: Please contact:
density with a sensor spacing of 20m Land seismic exploration: Stefano Savazzi, Umberto Spagnolini,
(still reasonable for seismic acquisi- DEI, Politecnico di Milano, Italy
tions) can make WiMedia feasible for E-mail:,
WGN applications with minimal modi- Wireless Geophone Network:
fications. On the other hand, higher geo-
phone densities, with a sensor spacing Vittorio Rampa
of less than 11m, require further MAC IEIIT-CNR, Italy
modifications. More details are given at WiMedia: E-mail:
the WisyGeo Web site.

A Software Platform for the Acquisition and

Online Processing of Images in a Camera Network
by Thomas Sarmis, Xenophon Zabulis and Antonis A. Argyros

Applications related to vision-based monitoring of spaces and to the visual understanding of human
behaviour, require the synchronous imaging of a scene from multiple views. We present the design and
implementation of a software platform that enables synchronous acquisition of images from a camera
network and supports their distribution across computers. Seamless and online delivery of acquired data
to multiple distributed processes facilitates the development of parallel applications. As a case study, we
describe the use of the platform in a vision system targeted at unobtrusive human-computer interaction.

Camera networks are increasingly

Application Layer
employed in a wide range of Computer
Vision applications, from modelling and
interpretation of individual human
Host computer Host computer
behaviour to the surveillance of wide
areas. In most cases, the evidence gath- Processing modules Processing modules
ered by individual cameras is fused
together, making the synchronization of
acquired images a crucial task. Cameras
Control messages

Local shared-memory Local shared-memory

are typically hosted on multiple comput- Cross-computer
ers in order to accommodate the large shared-memory
number of acquired images and provide
Acquisition module Acquisition module
the computational resources required for Synchronization
their processing. In the application
layer, vision processing is thus sup- Cameras Cameras
ported by multiple processing nodes
(CPUs, GPUs or DSPs). The proposed Figure 1: Platform architecture for camera network supported by multiple computers. Central
platform is able to handle the consider- sensor control is provided by a message-passing communication infrastructure. Acquisition
able technical complexity involved in modules place the images on a synchronized shared-memory space, making them available to
the synchronous acquisition of images multiple processing nodes across computers. Processing modules read these data and perform
and the allocation of processes to nodes. parallel execution of computer vision algorithms. The same modules also have access to this
Figure 1 illustrates an overview of the space, allowing them to synchronously view intermediate computation results.
proposed and implemented architecture.

The platform integrates the hardware and

device-dependent components employed System modules can communicate in bandwidth requirements imposed by
in synchronous multi-camera image and two modes. Communication through image transmission are accommodated
video acquisition. Pertinent functionali- message-passing addresses control by a Direct Memory Access channel to
ties become available to the applications messages to targeted or multicast recip- a local shared-memory space. For
programmer through conventional ients. The diversity of communicated cross-computer availability of images,
library calls. These include online control information types is accommodated by memory spaces are unified over a net-
of sensor-configuration parameters, data-structure serialization. Communi- work link. The latency introduced by
online delivery of synchronized data to cation through shared-memory spaces this link is compensated for by notifica-
multiple distributed processing nodes, provides visual data or intermediate tion of nodes, regarding the partial or
and support for the integration and sched- computation results to the nodes of the total availability of a synchronized
uling of third-party vision algorithms. host or of multiple computers. The large image set. In this way, per-frame syn-

ERCIM NEWS 76 January 2009 45

Special Theme: The Sensor Web

Figure 2: Person silhouettes in synchronous images

Image acquisition
are segmented in parallel. Individual processes fuse
them into a registered 3D representation of the per-
son, recognize coarse gestures as expressed by body
configuration, and estimate the spatial direction in
which the person's head is facing.

Acquisition module Acquisition module


Processing modules Processing modules

Cross-computer shared-memory

Processing module Processing module Processing module

3D representation Head pose Body configuration
map registration estimation analysis

chronization of modules is achieved, development is facilitated by support and analysis operations is performed in
but at the same time, processing of par- for 'chaining' of processes. parallel on each image, to detect the
tially available input is also supported. presence of humans through back-
Shared-memory spaces across process- Being in the format of a binary library, ground subtraction in the acquired
ing nodes are essential, as large data this platform can be invoked, independ- images. Using the shared memory
capacity and frequent input rate demand ent of the programming language used. across computers, segmentation results
the parallelization and pipelining of As an additional utility, the developed are fused into a 3D volumetric repre-
operations. platform provides a GUI for the control sentation of the person and registered to
of generic camera networks and the a map of the room. Two other processes
Acquisition modules encapsulate the recording of image sequences. Forth- run in parallel and access the same data
complexity of sensor-specific, syn- coming extensions involve additional to recognize the configuration of the
chronization, and shared-memory con- capabilities for cooperation with mid- person's body and estimate the pose of
figurations. Online sensor configura- dleware infrastructures in systems the person's head. The utilization of the
tion and command is implemented where vision is integrated with other proposed platform facilitates the modu-
through message-passing, while image sensory modalities (aural, tactile etc). lar development of such applications,
transmission utilizes shared-memory improves the reusability of algorithms
communication. A range of off-the- The platform is currently employed in and components and reduces substan-
shelf sensor types is supported through the development of a vision system tially the required development time.
an extensible repository of device-spe- (illustrated in Figure 2), targeting unob-
cific wrappers. To facilitate testing of trusive and natural user interaction. The This work has been partially supported by
applications, input may be prere- development of this system is part of a the FORTH-ICS RTD programme 'AmI:
corded. broader project funded internally at Ambient Intelligence Environments'.
FORTH-ICS on Ambient Intelligence
Processing modules run vision algo- (AmI) environments. The system Link:
rithms that are transparent to the com- employs multiple cameras that jointly
puter and provide access to images and image a wide room. Two computers doku.php?id=intro
intermediate computation results. Dur- host eight cameras and a dedicated bus
ing the applications development stage, for their cross-computer synchroniza- Please contact:
an Application Programming Interface tion, and utilize a LAN connection for Xenophon Zabulis
enables synchronization and message communication. Upon image acquisi- FORTH-ICS, Greece
coordination. Articulated application tion, a sequence of image processing E-mail:

46 ERCIM NEWS 76 January 2009

Tackling the Semantic Gap
in Multimodal Sensor Networks
by Eric Pauwels, Albert Salah and Paul de Zeeuw

Sensor networks are increasingly finding their way into our living environments, where they perform
a variety of tasks like surveillance, safety or resource monitoring. Progress in standardization and
communication protocols has made it possible to communicate and exchange data in an ad hoc
fashion, thus creating extended and heterogeneous multimodal sensor networks. CWI is looking at
ways to automatically propagate semantic information across sensor modalities.

Wireless sensors are deployed in a grow- Indeed, an additional layer of intelli- data picked up by complementary sen-
ing number of applications where they gence on top of the communication pro- sors (or modalities) are linked to
perform a wide variety of tasks. Although tocols will enable sensors to advertise semantically related concepts. A sim-
this has considerable economic and social their own capabilities, discover comple- ple example will clarify the issues at
advantages, it seems likely that even mentary services available on the net- hand: imagine a camera network on a
greater benefits can be gained once het- work and orchestrate them into more factory floor that has been programmed
erogeneous sets of individual sensors are powerful applications that meet high- to identify persons using face recogni-
able to communicate and link up into level specifications set by human super- tion, and to determine whether or not
larger multimodal sensor (inter)networks. visors. This can be achieved more effi- they are walking, say for safety rea-
We expect that the network's performance ciently if the capabilities of the different sons. If the same factory is also
will become more robust when informa- components can be described in both equipped with open microphones that
tion from multiple sources is integrated. human- and machine-readable form. It monitor ambient noise, then an intelli-
gent supervision system might pick up
the strong correlation between walking
people as observed by the camera net-
work and rhythmic background sounds
Figure 1: In multimodal sensor as detected by the microphones.
networks, reliable information
from one sensor can be used to By mining general knowledge data-
supervise the extraction of bases, the system might then be able to
semantic information from conclude that the observed rhythmic
another sensor. audio output corresponds to the sound
of footsteps and add this snippet of
semantic information to its knowledge
database. In essence, the system suc-
ceeded in using available high-level
information (the visual recognition of
walking people) to bridge the semantic
gap for an unrelated sensor (audio). By
accumulating the information gleaned
In addition, networks could become will then be possible for individual sen- from such incremental advances, we
smarter for at least two reasons: sensors sors to relate their own objectives and contend that it will be possible to grad-
that produce highly reliable output can be capabilities to human-defined goals (eg ually - but largely automatically -
used to provide on-the-fly 'ground truth' minimize energy consumption without extend the system's knowledge data-
for the training of other sensors within the sacrificing comfort) or available knowl- base linking low-level observed sensor
network, and correlations among sensed edge, both of which are usually data to high-level semantic notions.
events could bootstrap the automatic expressed in terms of high-level seman-
propagation of semantic information tics. To explore the viability of this idea we
across sensors or modalities. have conducted a number of simple
Granted, the linking of low-level sen- experiments in which we used the Inter-
The implementation of our vision sor data to high-level semantic con- net as a general knowledge database.
requires two conditions to be met. cepts remains a formidable problem, For instance, referring to the above sce-
Firstly, sensors should come equipped but we contend that the complementar- nario we submitted the paired search
with an open interface through which ity inherent in the different sensing terms walking (as the camera has been
their output data and all relevant meta- spectra supported by such a network programmed to detect this behaviour)
data can be made available for third might actually alleviate the problem. and sound (through the use of standards
party applications. Secondly, sensor net- The basic idea is simple: if particular such as SensorML, each sensor can
works need to be endowed with a learn- sensor data can be linked to specific communicate the modality of its output)
ing mechanism that shifts the burden of semantic notions, then it can be into a search engine and analysed the
supervision from humans to machines. hypothesized that strongly correlated response. By restricting attention to

ERCIM NEWS 76 January 2009 47

Special Theme: The Sensor Web

meaningful words that occur frequently their semantic distance to the original Link:
(both in terms of number per page and concept (walking). By restricting atten-
number of unique pages), we end up tion to the most similar concepts, it
with a sorted list that suggests a link transpires that it is highly likely the Please contact:
between the audio data and a list of recorder audio is related to either foot- Eric Pauwels
semantic concepts including music, steps, gait or music, all of which make CWI, The Netherlands
video, gait, work and footsteps. In a sense. These results hint at the possibil- Tel:+31 20 592 4225
final step this list is further whittled ity of automatically extending semantic E-mail:
down by checking each of these sugges- notions across modalities, thus leading
tions against an ontology to determine to more robust and intelligent networks.

Utilising Wearable Sensor Technology

to Provide Effective Memory Cues
by Aiden R. Doherty and Alan F. Smeaton

We describe a wearable sensor technology that passively records 'lifelog' images and sensor readings
of a wearer's daily life. The focus of our work is not on aggregating, collecting or networking data as in
the usual application of sensors in the Sensor Web, but rather on detecting events of interest to the
wearer from a multi-sensor standalone device. These events of interest provide effective cues to allow
people to more easily access their autobiographical memories. Early research indicates this
technology may be potentially helpful for sufferers of neurodegenerative diseases such as Alzheimer's.

Sensors and sensing technology are Lifelogging is the term used to potentially useful as a memory aid to
everywhere, and this issue of ERCIM describe the recording of different recall autobiographical memories.
News contains many examples of sen- aspects of your daily life, in digital Research in the field of cognitive neu-
sors networked together for some greater form, for your own exclusive personal ropsychology has established that 'cued
purpose. Mostly, people deploy sensors use. It can take many forms, such as an recall' is better than 'free recall'. The
and then gather the readings together and application running on your mobile closer a cue is to how an actual memory
address issues like networking, calibra- phone that 'logs' all your phone calls. was encoded, the better memory
tion, sensor fusion and sensor event One particularly interesting device is retrieval is. Other studies indicate that
detection. The general trend is towards the SenseCam, a camera that is worn autobiographical memories tend to be
networking sensors into the Sensor Web, around the neck and automatically strongly encoded in a visual manner in
but this isn't the only way of using them. captures thousands of images of the the brain. The SenseCam records pic-
Sensors can be used in small groupings wearer's life every day. It has a range tures from the viewpoint of the user,
on standalone devices that gather and of in-built sensors for monitoring the making it able to provide visual cues of
process information and report back not wearer's environment, detecting move- our past that are very close to how the
sensor readings, but major semantic ment, ambient temperature, passive original memories/experiences were
events. In this article we describe one infrared information (ie body heat) and encoded in the brain.
such sensor technology which is simple light intensity.
and cheap to manufacture, but can Even though SenseCam images provide
empower an individual to reflect on their Preliminary studies indicate that infor- strong memory cues, there exists a sub-
past behaviour and memories. mation gathered by the SenseCam is stantial problem in effectively manag-
ing the overwhelming volume of
images generated by this device –
approximately 650 000 images per year
are captured. Within the CLARITY
centre at Dublin City University, we
have developed a suite of functions
applied to SenseCam data that automat-
ically provide effective digital memory
retrieval cues. We structure our process-
ing into a number of logical steps that
exploit various characteristics of the
human memory system.

1. Firstly, we intelligently segment

sequences of images into distinct
Figure 1: Sample lifelog images from a typical IT worker's day. events such as having breakfast,

48 ERCIM NEWS 76 January 2009

working on a computer etc. This is detecting how visually novel each to gather information as part of a lifelog.
achieved very quickly using on-board event is. These include Universities of Toronto,
environmental sensor values. Tampere, Illinois, Utrecht and CWI in
4. As human memory is known to store Amsterdam. Our approach does not
2. Given that human memory stores items associatively, it is useful to aug- conform to the common model of a sen-
information associatively, we provide ment individuals' SenseCam events sor network composed of inter-con-
users with automated search func- with images (or videos) from external nected sensors with live, real-time
tions to find events similar to a given sources, eg to better remember a trip streaming data. This is because the
event, eg "show me other times when to the Eiffel Tower by viewing pic- demands of lifelogging are for post-
I was at the park". By intelligently tures of the tower uploaded by others event reflective retrieval rather than real
representing events through the to the Internet. Using GPS informa- time, meaning live inter-connectivity
fusion of image descriptions and the tion, and after some intelligent auto- with other sensor nodes is not as vital as
in-built sensor values, we found that mated processing, we can automati- in other Sensor Web technologies.
users can find events related to any cally find relevant supplementary
given 'query event'. images and videos from Internet sites Link:
such as Flickr and YouTube.
3. Given that the human memory more
strongly encodes distinctive memo- Our technology for autobiographical Please contact:
ries, we automatically identify events memory capture and management has Aiden R. Doherty, Alan F. Smeaton
that are more visually unique among been deployed not only within our CLARITY Centre for Sensor Web
those recorded by a wearer. We have research centre in Dublin, but also in Technologies, Dublin City University,
found that it is effective to combine numerous cognitive psychology Ireland
the automated detection of faces (to research groups in Europe and North E-mail:,
indicate social engagement) with America, and uses sensors and a camera

SENSE – Smart Embedded Network

of Sensing Entities
by Wolfgang Herzner

The SENSE project (Smart Embedded Network of Sensing Entities) is developing methods, tools and a
test platform for the design, implementation and operation of smart adaptive wireless networks of
stationary embedded sensing components. The network is an ambient intelligent system, which
adapts to its environment and delivers reliable information to its component sensors and the user.

SENSE is an EC-funded project of the tem consisting of multiple embedded • to understand how a shared semantic
6th Framework Programme, Embedded components: video system, audio sys- vocabulary influences dynamic node
Systems (objective 2.5.3, contract no. tem, central processor, power source discovery and configuration
33279). It aims at developing a platform and wireless networking. The security • to understand how perception and
for smart adaptive wireless networks of application implements object/scenario information processing can be com-
smart sensors. These sensors cooperate recognition (eg unattended luggage or bined using low-level and high-level
to establish and maintain a coherent people 'lurking' in an area). Nodes rec- feature fusion.
global view from local information. ognize local objects, using a combina-
Newly added nodes automatically cali- tion of video and audio information, and The expected results of SENSE are to
brate themselves to the environment, neighbouring nodes exchange informa- combine the aspects of:
and share knowledge with their neigh- tion about objects in a self-organizing • embedded intelligent middleware in
bours. The network is scalable due to the network. The result is a global overview smart devices,
local processing and sharing of informa- of current objects and events observed • adaptive configuration,
tion, and self-organizes based on the by the network (see Figure 1). • flexible cooperation (among devices),
physical placement of nodes. • high-level perception and adaptation
The five main objectives are: • dynamic networking in a common
As test platform for a civil security mon- • to build networked systems of framework of semantic knowledge
itoring system, a test application com- embedded components that can discovery and sharing.
posed of video cameras and micro- dynamically and automatically
phones, was chosen. The test platform reconfigure themselves The SENSE system encompasses
will be installed at the Krakow Balice • to convert low-level local informa- aspects including:
airport, to yield real data and perform- tion to semantic knowledge • construction of a modality-neutral
ance goals from a realistic test environ- • to use semantic-level knowledge for embedded test platform
ment. Each sensor is a stand-alone sys- network-centric computation • raw sensory processing

ERCIM NEWS 76 January 2009 49

Special Theme: The Sensor Web

• transformation of sensory data into

semantic knowledge
• communication between nodes to
produce a consistent world view
• sharing of knowledge between intel-
ligent nodes
• automatic recognition of unusual and
alarm situations
• communication between the intelli-
gent network and an operator, and
• automatic discovery and configura-
tion of new intelligent nodes.

Embedded systems in SENSE develop

their own semantic symbols based on
an analysis of their environment.
SENSE incorporates research from
machine learning to discover statistical
regularities in its environment, and
compresses these regularities into Figure 1: Civil Security demonstrator (Airport Krakow); frame-by-frame detection of objects.
informative semantic symbols. At the
local level, SENSE uses algorithms
such as 'expectation-maximization' to adaptive and device-independent. Net- works, connections are constantly cre-
optimize each node's set of semantic works that cover those challenges are ated and destroyed. This is called 'plug
symbols. Sharing of knowledge called ad hoc networks or self-organiz- and participate'.
between nodes is also a topic of ing networks. Their development is
research, both in distributed systems driven by the wireless community, but Vision and audio sensors were selected
and artificial intelligence. The SENSE some of their principles are also of to fit the application domain and also
system uses a mature algorithm called interest for wired networks. Less effort because they are complex enough to
'belief propagation'. This algorithm is required for their installation, initial- allow for significant advancement in
specifies how to share probability dis- ization and maintenance, and they dis- sensor processing technology. How-
tributions over semantic concepts play inherent fault tolerance and the ever, the framework designed will be
between nodes, such that a self-consis- possibility to save energy within the generic enough to accommodate a wide
tent world view results. Figure 2 illus- network. This is true for both wired and range of sensors. The middleware and
trates the architecture of SENSE. wireless types but is typically only rele- software framework is designed to eas-
vant for the latter. In contrast to self- ily incorporate additional sensor types,
The unique feature of SENSE is that it organizing networks, traditional net- which mean the project results are more
combines technology from embedded works have a very time-demanding easily extended and the technology is
systems, robotics, networking and commissioning phase that also involves easier to adopt for third parties.
machine learning research in a new expert knowledge.
way. The result is a framework for the With respect to requirements, the air-
development of smart networks of Dynamic addition or removal of nodes port application is very similar to other
embedded components that are flexible, is a further challenge. In ad hoc net- applications in public buildings and
public areas, such as shopping centres,
railway and bus stations, office build-
ings, football stadiums and so on. The
SENSE technology is modular and is
appropriate for integration with existing
Semantic Co-activated is communicated security systems (existing video sys-
Semantic Symbols
information is to the user tems, for example). This provides an
communicated lead to shared
representations upgrade path from current systems to a
around the
network User full distributed SENSE system, improv-
Interface ing the chances for technology uptake.
Nodes build up
an understanding
Node Node
of their environment
by fusing sensor
input to form
semantic symbols Please contact:
Nodes perceive Objects
and Wolfgang Herzner
objects and
events in their Events Austrian Research Centers – ARC
environment Figure 2: (AARIT)
(video and audio) SENSE architecture. E-mail:

50 ERCIM NEWS 76 January 2009

Towards Data Management in the Sensor Web:
the MaD-WiSe System
by Giuseppe Amato, Stefano Chessa, Francesco Furfari, Stefano Lenzi, Claudio Vairo

The convergence of sensor networks with the Web (Sensor Web) poses new problems. These relate
both to the management of the enormous amounts of data continuously produced by sensors, and
to the reaction to events inferred from such data. The MaD-WiSe system (MAnagement of Data in
WIreless SEnsor networks), developed at ISTI-CNR, exploits the well-known database paradigm to
address this issue.

The MaD-WiSe system considers a

Wireless Sensor Network (WSN) as a
highly distributed and dynamic data-
base. Sensor data can be acquired,
manipulated and filtered using simple
SQL-like statements. Various WSN
applications can be developed using its
functionality. For example, the MaD-
WiSe system has been used to imple-
ment a prototype application providing
remote monitoring of firefighters
equipped with totally encapsulated
chemical suits (see Figure 1). This
application enables real-time monitor-
ing of various physiological parameters
during operational activity. The infor-
mation acquired can be used to raise the
alarm in situations of risk, provide
advice to the team leader with respect to
the necessary actions, and also to
analyse offline the health of those
involved in an operation. The applica- Figure 1: Firefighters being monitored with MaD-WiSe.
tion employs five sensors that are
deployed on the arms, legs and chest of
the operator.

The MaD-WiSe architecture comprises munications between nodes and data with both sensor and remote streams.
a set of modules running on WSN nodes processing. For the former, data rates determine the
(network side), and a set of modules that activation frequency of transducers
offer WSN services to the external The Stream System defines three types associated with sensor streams. In the
applications (context information of streams: sensor, remote and local latter case, data rates are used by the
provider). The network side consists of streams. A sensor stream is connected network layer to optimize radio sched-
a set of modules that implement a dis- to a transducer and carries data originat- uling: the radio is switched on only
tributed data stream management sys- ing from the transducer. A remote when a piece of data must be sent
tem on a WSN. It is organized into three stream is a data channel between two through a remote stream. Sensor
layers, as shown in Figure 2. The layers distinct sensors: writing to a remote streams can also be 'on demand'. In this
interact through well-defined interfaces stream occurs on one sensor while read- case the transducers are activated only
and are autonomous with respect to each ing from the stream occurs on the other. in response to an explicit read request
other. Thus remote communication between on the stream.
different sensors is encapsulated within
The Network Layer supports connec- the stream system, which in this respect The last of the three layers is the Query
tion-oriented multi-hop communication offers the transport layer functionali- Processor Layer, which implements the
between arbitrary pairs of nodes. The ties. A local stream is local to a sensor, query processor of a full-in-network dis-
Stream System Layer offers abstraction as writing to and reading from the tributed data stream management sys-
mechanisms for data access by means stream can only be requested by code tem. It can be programmed by the client-
of data streams. It can be thought of as running on the same sensor. side subsystem in order to take part in
the equivalent of a file system on a sen- The Stream System allows streams to the execution of a distributed query. The
sor network. The main difference is that be created or removed, and records to query language used in MaD-WiSe is
in the Stream System, data is continu- be read from and written to existing called MW-SQL and shares its basic
ously acquired from transducers, com- streams. Data rates can be associated constructs with SQL. However, sensor

ERCIM NEWS 76 January 2009 51

Special Theme: The Sensor Web

Wireless Sensor Network


7 6

MaD-WiSe Network-side node

Query Manager/Executor

Stream System



Figure 2: The software in a network-side node. Figure 3: The MadWise query interface.

network peculiarities and the distributive module (composed of a query parser, 2.x. It runs on WSN platforms based on
nature of the database implementation an execution plan optimizer and a MICAz and IRIS motes and is distrib-
introduce some differences. MW-SQL query manager) and a higher-level uted with an open-source licence. Fur-
allows users to express queries to manip- module, the JDBC driver, which inter- ther information and downloads can be
ulate, filter and organize sequences of acts with the low-level module by found on the project Web site.
tuples generated by the sensors. MW- means of the MW-SQL language. At
SQL relies on the concept of source to the current stage of the project, the
present the user with an abstraction of a JDBC driver is being encapsulated
sequence of tuples arriving from a pre- within an OSGi bundle in order to Link:
cise origin. The MaD-WiSe query inter- implement a gateway between the
face is shown in Figure 3. WSN and the Sensor Web, and to
enable queries involving different Please contact:
The MaD-WiSe context information WSNs in the Sensor Web. Giuseppe Amato
provider fits within the raw data ISTI-CNR, Italy
retrieval layer of a context-aware MaD-WiSe was developed using nesC Tel: +39 050 3152906
architecture. It comprises a low-level and is available for both TinyOs 1.x and E-mail:

Detecting Hazardous Gases in Emergency

Disaster Scenarios using Wearable Sensors
by Tanja Radu, Cormac Fay, King Tong Lau and Dermot Diamond

The aim of this project is the development of integrated smart wearable sensors for emergency
disaster intervention personnel. The CLARITY (The Centre for Sensor Web Technologies at Dublin City
University) team is involved in the integration of gas sensors into wearables for detection of
hazardous gases like CO and CO2.

This research arose through involve- for potential risk sources. Some of the projects (see for example www.biotex-
ment in a joint European Union-funded issues covered are monitoring of vital
FP6 project called Proetex (www.proe- signs, posture and activity, external The aim of the project is to hazard monitoring, and low-power The project brings together 23 partners
develop textile- and fibre-based inte- wireless communications. The project from a range of backgrounds – univer-
grated wearable sensor systems. Such is closely connected with other large sities, research institutions, industrial
systems will improve the safety and European projects on smart textiles, partners and end users – drawn from
efficiency of emergency personnel by wearable sensing and associated appli- laboratories in France, United King-
monitoring the health status of the oper- cations – it brings together and extends dom, Poland, Italy, Belgium, Switzer-
ator and the surrounding environment the technology developed by previous land, Germany and Ireland.

52 ERCIM NEWS 76 January 2009

Researchers from CLARITY based in ted wirelessly to the wearable base sta-
the National Centre for Sensor Research tion using Zigbee. Power is supplied to
at Dublin City University are involved the sensors using a nickel metal hydride
in the integration of sensing platforms rechargeable battery. The CO2 sensor is
into wearables for the detection of envi- placed in a specially designed pocket
ronmentally harmful gases surrounding located on the firefighter's boot. The
emergency personnel. Special attention pocket is designed not to obstruct the
is being paid to carbon monoxide (CO) firefighter's activities. The prototype
and carbon dioxide (CO2). These gases currently used for testing is shown in
are associated with fires and mining Figure 1; note the side pocket contain-
operations, and it is of the highest ing the CO2 sensor along with the wire-
importance to warn and protect opera- less sensing module and a battery. The
tors from potential harm caused by pocket has a waterproof membrane that
over-exposure to high concentrations of protects the sensor from moisture, but
these gases. The objective is rapid allows gas to pass through. The CO sen-
detection of the status of an environ- sor will be integrated in the firefighter's
ment (low, medium or high hazard) and outer garment (ie jacket). All sensed
real-time communication of this infor- Figure 1: Firefighter's boot with built-in information will be fed to a wearable
mation to the garment wearer. Critical pocket used for enclosing the CO2 sensor and local base station that shares the data
in this identification of potential toxifi- wireless communications platform. with a remote centralized base station.
cation is a reliable method of measuring The ultimate goal is to achieve local
CO/CO2 exposure. Commercially communication between firefighters and
available sensors have been carefully civil workers in the operations area, as
selected and are being integrated into in which the current between the elec- well as longer-range communications
the outer garments of firefighters. The trodes is proportional to the concentra- between these personnel and the support
sensors provide sufficient sensitivity to tion of the gas. On the other hand, the team outside the operations area.
reliably alert users to the presence of CO2 sensor is potentiometric. In this
these harmful gases. Another important case, the reference and working elec- The project commenced in 2006 and
aim is to achieve wireless transmission trodes are placed in an electrolyte that will end in early 2010. The project is
of sensor signals to a wearable wireless provides a reference CO2 concentration. envisioned to produce three sets of pro-
base station that gathers, processes and The measured potential is based on the totypes during the four years of its exis-
further transmits the data. difference in concentration between the tence; so far two generations of proto-
reference electrode and the outside air. types have been successfully developed.
When selecting the appropriate com- Both types of sensors are very sensitive At this stage, accurate wireless trans-
mercially available sensors for the gas and give an accurate reading (in parts mission of the sensor signal has been
sensing application, special attention per million). This means that both low successfully achieved (see Figure 2).
was paid to sensor size, robustness, sen- concentrations of these gases (which Future activities will include evaluation
sitivity and power requirement. Electro- can be hazardous over long periods of of prototypes in laboratory and field
chemical sensors satisfy most of these exposure) and high concentrations conditions. Their performance will be
requirements, especially in terms of (which pose an immediate danger) can compared to that of existing technology,
size and power requirements. CO is be accurately detected. The signal and will be improved upon by customiz-
detected using an amperometric sensor obtained from these sensors is transmit- ing the products according to the spe-
cific user needs. Finally, the products
will be tested in real-life situations as an
ultimate proof of their full functionality.

The authors gratefully acknowledge the

financial support of the European
Union (Proetex FP6-2004-IST-4) and
the Science Foundation Ireland


Please contact:
Tanja Radu
CLARITY Centre for Sensor Web
Technologies, Dublin City University
Figure 2: Wirelessly transmitted signal from CO2 sensor calibration (range atmospheric to Tel: +353 1 700 7602
42000 ppm CO2). Sensor was enclosed in an airtight chamber and CO2 was injected. E-mail:

ERCIM NEWS 76 January 2009 53

Special Theme: The Sensor Web

A Multi-Modal Sensing Platform for Sport
by Noel E. O'Connor, Philip Kelly, Ciarán Ó'Conaire, Damien Connaghan, Alan F. Smeaton,
Brian Caulfield, Dermot Diamond and Niall Moynahan

The evolution of the World-Wide Web to the Sensor Web is providing an unprecedented opportunity
to develop novel applications in a variety of domains. In this article we describe our work on multi-
modal sensing for sport.

Sensor technology is rapidly changing strains that they are feeling in their player speed using the location infor-
the professional sporting landscape. joints and muscles. mation and detecting the sound of a ball
Modern motor racing has been virtually hitting a racquet. These detected events
transformed by the introduction of The project is a collaboration between are used in a variety of Web-based
sophisticated drive-by-wire sensor tech- CLARITY (The Centre for Sensor Web coaching tools. These include (a) being
nology, for example. Similarly, the abil- technologies) and Tennis Ireland, the able to view individual tennis strokes
ity to accurately monitor the perform- national governing body for tennis, from multiple view points, (b) support
ance of an athlete during training is based in Dublin City University. We for online coaching feedback (eg “This
having a major influence on a wide have instrumented an all-weather tennis shot was the wrong choice given the
range of track and field events. In this court with nine Internet-enabled cam- location of your opponent when you
project we introduce state-of-the art eras with built-in microphones. This is played it”), and (c) an easy way for
sensing technology onto the tennis linked to a localization system that coaches and athletes to produce person-
court with a view to facilitating coaches identifies the player's position to within alized annotated video summaries for
as they train the next generation of ten- 15cm by triangulating the radio signal players to download and review for
nis superstars. These sensors make it emitted by small tags carried by the motivational purposes.

Integrating Smart Materials and Body

Sensor Networks
By its very nature, this project requires
convergence between multiple disci-
plines in order to be successful. It
requires input from engineers for signal
processing and content analysis aspects
and from computer scientists for
addressing indexing, archival, personal-
ization and user interface issues. Sports
scientists play the crucial role of medi-
ating between the technologists and the
end-user coaches and athletes. Their
input is invaluable in helping translate a
complicated set of domain-specific and
expert-driven requirements into a set of
concrete technical functionalities. This
helps ensure the practical relevance and
usefulness of any technology devel-
oped. They also help ensure that
coaches and athletes understand the
potential benefits of the technology,
Figure 1: Coaching tool developed for Tennis Ireland. thereby stimulating take-up by the key

possible for coaches to obtain a second- players in their pockets. The coach uses However, the potential for convergence
by-second record of player performance a simple wireless device to signal an extends even beyond this initial con-
that goes far beyond what can be cap- important event during play via a sim- stituency. More generally, the sensed
tured by more traditional techniques; ple button press. After training, these environment can be considered as an
simply eyeballing the player as they button presses are synchronized to the experimental platform for trialling
move, serve, volley and return, does not video streams. Content analysis mecha- wearable sensing. The wireless inertial
reveal what might be going on under nisms are then used to define the pre- sensing platforms being developed by
the surface in terms of the player's ever- cise start and end of tennis play around CLARITY engineers will be integrated
changing performance profile, his/her these locations. We do this by tracking into the system with a view to providing
breathing patterns, and the stresses and the ball in each camera, measuring biomechanical feedback to athletes.

54 ERCIM NEWS 76 January 2009

The smart materials being developed by built into the wearable platform of one of polyamide lycra. The device consists of
chemists in the CLARITY work pro- our industry partners and trialled with a super-absorbent material that provides
gramme will be integrated into body tennis players to monitor specific muscle a passive pumping mechanism to con-
sensor networks and will allow a variety movement and strain. Data collection trol fluid flow. The optical detection of
of important physiological and biomet- and feedback can be piggy-backed on pH-induced colour changes in the dye is
ric indicators to be detected. the existing system, with the added ben- achieved via a paired emitter-detector
efit of a synchronized multi-view video LED system. The Tennis Ireland instal-
To date, most data gathering and experi- to augment expert analysis of data from lation provides an ideal framework for
mentation with this technology happens the wearable sensors. testing this innovative technology in a
within a laboratory. The Tennis Ireland real application scenario with real users.
installation provides infrastructure for A longer-term example is the 'sweat
data gathering, analysis and feedback patch' being developed to measure the Please contact:
that will allow this technology to be pH of sweat. A change in sweat pH may Noel E O'Connor
tested in the field, during real exercise. signal an increased reliance on anaero- CLARITY Centre for Sensor Web
For example, the work on conducting bic metabolism, but there is a lack of Technologies, Dublin City University,
polymers will result in textiles for meas- reliable methods for assessing pH dur- Ireland
uring stretching, bending and pressure ing exercise. CLARITY is working on a Tel: +353 1 700 5078
movements. This 'smart foam' will be new fluid-handling platform based on E-mail:

Synchronizing Sensed Data

in Team Sports
by Dónall McCann, Mark Roantree, Niall Moyna and Michael Whelan

In this article we will be discussing the synchronization of sensor data in team sports.
Synchronization allows us to use more expressive queries, to query across all participants in a
given activity and to potentially discover new knowledge from the semantically enriched data.
A collaborative research effort between groups working on data management and on health
and human performance (both at Dublin City University) involved a series of experiments using
wearable sensors during team games and the capture and querying of sensed data.

When dealing with sensor data for a that correspond to various states, eg increasing activity through Pre-Game
team sport, it is often useful to be able to first half, second half etc. A 'profile' is a and Warm-Up, and remaining con-
query across multiple sensors and thus combination of various states. Each stantly active throughout each half. This
to be able to compare data from several state occurs once and in the order spec- profile can be easily split into states
players for any given moment in time. ified. The goal is to semantically enrich because of the period of rest located
In order to do this, the data from all sen- sensor data with an additional field that between the two periods of high activ-
sors must be synchronized so that the identifies the state associated with ity. However, this profile is atypical
start time of the game or activity can be every sensor reading. Our method is to among the thirty players involved in a
identified in the data from each individ- convert the sensor stream to XML, given game. A more typical graph is
ual sensor. which facilitates the subsequent seman- shown in Figure 2, corresponding to a
tic enrichment process. In simple terms, defensive player. This graph is charac-
This is necessary because sensor devices the synchronization process involves terized by short bursts of activity inter-
may be activated asynchronously, since identifying one or more specific spersed with periods of rest, making it
the device begins recording when it first moments in time, such as the beginning much more difficult to correctly iden-
comes into contact with the player's or end of the game. Once the reading tify the state boundaries. This provides
skin. While many sensor devices will corresponding to that time is identified, a significant challenge to creating a
record a start time, this information is the data can be synchronized with the generic process for normalizing and
not necessarily reliable as there is often data from all the other devices involved synchronizing sensor streams.
no correlation between the system time in the experiment.
and the time kept by the match officials, In order to perform our synchroniza-
or indeed between the times on any two The sensors used in our experiments tion, we define a 'model' profile of the
sensors. In addition, the devices may be record a heart rate value every 5 sec- ideal shape of the data graph. This com-
unreliable and may malfunction, or the onds, and approximately 1200 values prises two periods of consistently high
device may become detached during the are generated while the device is worn. activity on either side of a period of rel-
course of the game. The six states corresponding to a Gaelic atively low activity. This model profile
football match can be seen in Figure 1. is compared to the data from each sen-
From an abstract perspective, sensors This example graph is for a midfield sor device until the closest match is
can be regarded as generating values player who has a profile of gradually found, in terms of intensity and dura-

ERCIM NEWS 76 January 2009 55

Special Theme: The Sensor Web

Figure 1: Midfielder data with distinct state boundaries. Figure 2: Defensive player data with blurred state boundaries.

tion of the activity. Identifying state to identify at least one, namely the
boundaries for the resulting profile is beginning of the half-time period,
reasonably simple, involving identify- when every player ceased intense
ing the point at which the data changes activity and so their heart rates
in intensity and applying some rules dropped significantly. Once this point
from the domain experts to find the pre- is identified, the state durations Link:
cise location on the curve at which the extracted from the model profile,
state changes. The durations of each which should correspond to the dura-
state are then recorded based on these tions of the game periods, are applied Please contact:
boundaries. and each sensor reading is marked up Dónall McCann, Mark Roantree,
with state information. Niall Moyna, Michael Whelan
For the data from the other sensor Dublin City University, Ireland
devices, a single point is identified. In When all readings are assigned a state, E-mail:
our experiments, we discovered that it is possible to query the data according,
while in many cases it proved impossi- to state and across multiple sensors,,
ble to find precise state boundaries for resulting in richer knowledge and the,
all states, in every case it was possible potential for new knowledge discovery.

Inertial Sensing: A Little Bit of CLARITY

by John Barton, Brian Caulfield & Niall Moyna

The increasing availability of cheap, robust and deployable sensor technology will usher in a new
wave of ubiquitous information sources. A particular implementation of ambient sensors is in the
area of wearable electronics in body area networks incorporating inertial sensing devices. As part of
the CLARITY Centre for Sensor Web Technologies, the Tyndall Wireless Inertial Measurement Unit
(WIMU) is being used in a number of projects focussing on two key themes: Health and Fitness, and
Helping the Aged.

Wireless Inertial Measurement Unit a fixed frame of reference for position 25mm wireless node has been used to
The Tyndall Wireless Inertial Measure- measurement (x, y, z), the Earth-Fixed develop a platform for low-volume
ment Unit (WIMU) is a 6 Degrees of Frame, and utilizing a moving non- prototyping and research in the wire-
Freedom (6DOF) inertial sensing inertial frame (u, v, w), the IMU-Fixed less sensor network domain. A number
device, comprising triple-axis accelero- Frame, which has its axes parallel to of research projects currently under-
meters, gyroscopes (angular velocity) those of the IMU sensors. way at the institute are using it as a
and magnetometers. The triple-axis platform for sensing and actuating in
acceleration and angular velocity sen- The 25mm WIMU was developed scalable, reconfigurable distributed
sor output values can be combined in a based upon Tyndall's 25mm modular autonomous sensing networks, and it
nonlinear matrix equation to give both wireless sensor node technology. It is is supported by Science Foundation
position and orientation information. one of a large family of layers cur- Ireland (SFI) through Tyndall's
The system can be visualized by using rently available for the Tyndall25. The National Access Program (NAP).

56 ERCIM NEWS 76 January 2009

CLARITY - Centre for Sensor Web Technologies
The CLARITY Cenre for Sensor Web Technology focuses from the physical world in which we live and the digital
on the intersection between two important research areas – world of modern communications and computing. CLAR-
adaptive sensing and information discovery – to develop ITY commenced in June 2008 and represents a large-scale
innovative technologies of critical importance to Ireland's academia-industry collaboration, accommodating more
future and to improving quality of life in areas such as per- than 100 full-time researchers from University College
sonal health, digital media and environmental manage- Dublin, Dublin City University and the Tyndall National
ment. The theme of CLARITY's research programme – Institute, in partnership with more than ten industrial part-
bringing information to life – refers to the harvesting and ners, including major multinationals and emerging Irish
harnessing of large volumes of sensed information, both companies.

Inertial Sensing for Health for individuals across their lifespan. In field, to examine for example the 'cut-
and Fitness addition, the technology has the poten- ting' movement of players when they
The development of unobtrusive sens- tial to allow sports coaches and trainers make a rapid change in direction and
ing elements embedded in the fabric of to monitor individual athletes in a train- the stresses and strains that these
garments has opened countless possibil- ing or competitive environment. Cur- intense motions put on their joints and
ities for the innocuous monitoring of rently, coaches/trainers are very limited muscles.
athletes over extended periods of time in in what they can measure in real time in
a variety of sport settings. Foster Miller a training or competitive environment. As part of a collaboration with Tennis
(an independent company but part of the The information from the proposed sen- Ireland, tennis players will be fitted
QinetiQ Inc. group) has recently devel- sor platform could be used to design with WIMUs to augment the already
oped a T-shirt-based Ambulatory Physi- training programs that replicate the rich sensory environment available at
ological Monitoring System, which movement patterns and/or physiologi- the tennis facilities at Dublin City Uni-
monitors the vital signs of a person dur- cal responses for a given sport or for a versity. An all-weather tennis court has
ing activity and transmits the data wire- specific position (eg fullback vs. centre been instrumented with nine Internet-
lessly to a remote station. forward) in a team sport. enabled cameras with built-in micro-
phones. Adding WIMUs to the tennis
The combination of textile sensors with For biomechanical analysis, the ability players' bodies will enable us to deter-
WIMUs will greatly assist in the ambu- to monitor athletes' movement in their mine the actions they are performing
latory monitoring of healthy individuals natural environment is a huge leap for- and even the stroke they are playing.
and of those with chronic diseases such ward compared to the current method of
as obesity, diabetes, heart failure, and measuring them in a laboratory setting. Links:
arthritis. The information will allow As part of CLARITY, critical markers
patients and allied health professionals such as the speed and agility of top-
to monitor physiological response dur- level rugby players will be determined
ing various forms of activity, and to in the lab. They will then be outfitted
design individually tailored programs with WIMUs for assessment in the

Please contact:
John Barton
Tyndall National Institute, Cork,
Tel: +353 21 4904088

Brian Caulfield,
University College Dublin, Ireland
Figure 1: Tyndall Wireless Iner- Tel: +353 1 7166502
tial Measurement Unit showing E-mail:
yaw, pitch and roll.
Niall Moyna,
Dublin City University, Ireland
Tel: +353 1 7008802

ERCIM NEWS 76 January 2009 57