Annual Report

Institut für Parallele und
Verteilte Systeme
Annual Report
2009/2010
www.ipvs.uni-stuttgart.de
Institute of Parallel and
Distributed Systems

11
Preface
Celebrating the 20th anniversary of our institute was definitely the most pleasant
event in the period covered by this report. On June 19, 2009 we organized a collo-
quium celebrating 20 years of exciting research at IPVS. We had the pleasure to intro-
duce three outstanding researchers to give talks at this colloquium: Dr. J. Cheng (IBM
Fellow und Vice President, IBM Almaden Research Center, USA), Prof. T. Fukuda
(Dep. of Micro-Nano Systems Engineering, Nagoya University, Japan) and Prof. U.
Ramachandran (College of Computing, Georgia Tech, USA). Many thanks for their
most inspiring talks! We also would like to thank for the kind addresses given at the
colloquium, in particular for those of Prof. A. Reuter, the founder of the institute, and
Dr. H. Knorr, who supported the development of IPVS over the years.
While in 2008 our robot soccer team took a remarkable fourth place in the World
Championship, it was the winning team of both the Robocup World Championship
2009 and Robocup German Open in 2009. We congratulate the Roboter Fußball
Club (1. RFC) Stuttgart and his coach Prof. Levi on this excellent performance.
During the past year members of IPVS (co-)organized a number of scientific
events, including International Workshop on Quality of Context (QuaCon), ITPC
(IBM Technology Partnership Center) Colloquium on Cloud Computing and Nexus
Industry Colloquium, which all attracted more than 100 participants.
Board of Directors:
(from left)
Dipl.-Inf. Michael
Matthiesen (IFS),
Prof. Dr. Bernhard
Mitschang (AS),
Prof. Dr. Sven Simon
(PaS),
Prof. Dr. Kurt
Rothermel (VS; Head
of Institute),
Prof. Dr. Paul Levi
(BV),
Prof. Dr. Marc
Alexander Schweitzer
(SgS)

2
In October 2009, Prof. Mitschang was awarded the IBM Faculty Award for his
work on “Acceleration Technologies for Interactive Advanced Analytics to Support
New Intelligence for a Smarter Planet”. The IBM Faculty Award is a highly recognized
international award for excellent achievements in science and teaching as well as for
building a resilient cooperation between research and industry.
Two new colleagues joined IPVS, both strengthening the field of simulation. On
May 15, 2010 Prof. M. A. Schweitzer became head of the chair “Simulation of Large
Systems”. His research is in the field of Scientific Computing. Prof. M. Bader took
over the Juniorprofessor position “Simulation Software Engineering” on December 1,
2009. He is conducting research in the area of efficient algorithms and software devel-
opment in scientific computing. We warmly welcome both of them and wish them the
very best for their work at IPVS.
Finally, we would like to thank our partners and friends in industries and aca-
demia for their continuous support. We also appreciated the great effort of our staff to
bring IPVS forward!
For the Board of Directors: Prof. Dr. rer. nat. Dr. h.c. Kurt Rothermel
Head of the Institute IPVS

Annual Report 2009/2010
(April 1, 2009 - March 31, 2010)
Universität Stuttgart
Institut für Parallele und Verteilte Systeme (IPVS)
Universitätsstraße 38
70569 Stuttgart
Deutschland
www.ipvs.uni-stuttgart.de
Published by: Board of Directors of the IPVS
Edited by: Ute Gräter
Michael Matthiesen
Stefanie Palmer
Design by: Michael Matthiesen
Every reproduction - even in extracts - only with the
written consent of the publisher.
© IPVS, 2010

55
Content
Preface 1
Content 5
1 Organisational Structure 7
2 Research Activities 9
2.1 Applications of Parallel and Distributed Systems 9
2.2 Applied Computer Science - Image Understanding 17
2.3 Parallel Systems 23
2.4 Simulation of Large Systems 30
2.5 Distributed Systems 31
2.6 Joint Activities 40
2.7 Infrastructure 42
3 Teaching 45
4 Dissertations and
Postdoctoral Qualiﬁcations 47
Dissertations
5 Publications 55
Books / Book Chapters
Articles in Journals
Articles in Proceedings
Others
6 Events 65
7 Professional Activities 69
Memberships
Program Committees

6
Organisational
Structure
Management
Head of Institute
▶ Prof. Dr. rer. nat. Dr. h.c. Kurt RothermelR
Deputy Head of Institute
▶ Prof. Dr.-Ing. habil. Bernhard Mitschangrd Mitsch
Board of Directors
▶ Prof. Dr. rer. nat. habil. Paul Levievi
▶ Dipl.-Inf. Michael Matthiesen
▶ Prof. Dr.-Ing. habil. Bernhard Mitschangits
▶ Prof. Dr. rer. nat. Dr. h.c. Kurt RothermelRother
▶ Prof. Dr. rer. nat. Marc Alexander Schweitzerxander Schwe
▶ Prof. Dr.-Ing. Sven Simon

77
1 Organisational Structure
Departments
Applications of Parallel and Distributed Systems
Anwendersoftware (AS)
▶ Prof. Dr.-Ing. habil. Bernhard Mitschang
email: Bernhard.Mitschang@ipvs.uni-stuttgart.de
Tel: +49-711-7816449 - Fax: +49-711-7816424
Applied Computer Science - Image Understanding
Praktische Informatik - Bildverstehen (BV)
▶ Prof. Dr. rer. nat. habil. Paul Levi
email: Paul.Levi@ipvs.uni-stuttgart.de
Tel: +49-711-7816387 - Fax: +49-711-7816250
Parallel Systems
Parallele Systeme (PaS)
▶ Prof. Dr.-Ing. Sven Simon
email: Sven.Simon@ipvs.uni-stuttgart.de
Tel: +49-711-7816450 - Fax: +49-711-7816250
Simulation of Large Systems
Simulation großer Systeme (SgS)
▶ Prof. Dr. rer. nat. Marc Alexander Schweitzer
email: Marc.Alexander.Schweitzer@ipvs.uni-stuttgart.de
Tel: +49-711-7816465 - Fax: +49-711-7816248
Distributed Systems
Verteilte Systeme (VS)
▶ Prof. Dr. rer. nat. Dr. h.c. Kurt Rothermel
email: Kurt.Rothermel@ipvs.uni-stuttgart.de
Tel: +49-711-7816434 - Fax: +49-711-7816424
Infrastructure
Infrastruktur (IFS)
▶ Dipl.Inf. Michael Matthiesen
email: Michael.Matthiesen@ipvs.uni-stuttgart.de
Tel: +49-711-7816414 - Fax: +49-711-7816248

99
2 Research Activities
2.1 Applications of Parallel and
Distributed Systems
Research Focus
The importance of information in business and also in our daily life has drastically
increased in the last few years. We are constantly challenged by escalating variety, ve-
locity, and volume of data. Our understanding of information management is about
creating and enabling business by integrating, analyzing, optimizing, and storing het-
erogeneous types and sources of information throughout its life-cycle and in an ef-
fective manner. This is reflected in the different research projects that are conducted
in our department of Applications of Parallel and Distributed Systems. The research
work of the department can be categorized into the following three working areas:
▶ Data and Metadata
Fundamental prerequisites for any effective information management comprise
efficient data and metadata management by means of enhanced database tech-
nology. Explicit metadata management further enables a semantic treatment, in-
teroperability as well as a code generating approach as applied by model-driven
engineering.
▶ Content and Semantics
Content management defines the complementing facet of information manage-
ment. Content is understood to be unstructured or semi-structured and thus de-
mands for different storage, search, and management techniques as compared to
Universität Stuttgart - IPVS
Anwendersoftware
70569 Stuttgart
Germany
Prof. Dr.-Ing. habil. Bernhard Mitschang
 Bernhard.Mitschang@ipvs.uni-stuttgart.de
 +49-711-7816449
 +49-711-7816424

10
the well-structured data management area. Approaches that bridge this gap try to
address both aspects in order to get the best of both sides.
▶ Information Systems, Applications, and Services
The above mentioned technologies enable the development of new techniques
for information systems and associated applications, like process or product data
management, message monitoring and retention systems, or context-based infor-
mation systems. The service-based approach is seen as a next technology step to-
wards enhancing flexibility.
The research activities of the department are best characterized by its current involve-
ment in research projects as described further in the project description section below.
The department is involved in two programs as part of the excellence initiative granted
to the Universität Stuttgart by the German Research Foundation (DFG):
▶ The department coordinates all activities within the topical cluster “Information
and Communication Technologies for Manufacturing” that is part of the Gradu-
ate School GSaME (Graduate School advanced Manufacturing Engineering).
▶ The department participates in the research area “Integrated Data Management
and Interactive Visualisation” as part of the Cluster of Excellence “Simulation
Technology” (SimTech).
In 2009 Prof. Mitschang was awarded the prestigious IBM Faculty Award for his work
on “Acceleration Technologies for Interactive Advanced Analytics to Support New In-
telligence for a Smarter Planet”.

11
Staff
Research Projects
www.ipvs.uni-stuttgart.de/abteilungen/as/forschung/projekte
DaMT - Data and Multi-Tenancy
Personnel: Oliver Schiller
Funded by: Universität Stuttgart
Duration: since 1.1.2009
Multi-Tenancy is about providing a Software as a Service offering for multiple tenants
in a cost-efficient way. Thereby, an accomplished manageability of the service and a
clear-cut and secure isolation between tenants constitute the key challenges. In this
project, we focus on extending traditional data management concepts and systems by
Multi-Tenancy.
Staff of the Department:
(from left) V. Rewucki, M. Großmann, C. Lübbe, O. Schiller, A. Brodt, A. Roesler, H. Schwarz,
S. Radeschütz, P. Janowski, J. Minguez, T. Scheibler, Ch. Stach, P. Reimann, S. Silcher, B. Mitschang,
F. Niedermann, N. Cipriani
(staff not shown on the photo) A. Haufler, M. Haufler, N. Hönle, M. Jakob, A. Moosbrugger, M. Vrhovnik,
F. Wagner

12
BIA - Business Impact Analysis
Personnel: S. Radeschütz
Funded By: IPVS, partially Deutsche Forschungsgemeinschaft (DFG)
For optimizing business processes effectively, a profound analysis of all relevant busi-
ness data in a company is necessary. This project aims to develop an approach to auto-
matically integrate a company’s process execution data and operational business data
in order to enable a more informative and comprehensive analysis. The BIA project
addresses important issues like the development of matching algorithms, of extended
analysis techniques in OLAP and data mining and of process optimization procedures.
GSaME – Cluster E - Data Management and Information Provision-
ing in the Virtual Factory
Personnel: S. Silcher, J. Mínguez, M. Jakob, T. Scheibler
Funded By: Graduate School advanced Manufacturing Engineering (GSaME)
Duration: 1.4.2008 - 30.9.2013
The real-time factory offers an intelligent, real-time operational management of facto-
ry processes and resources. The shift to real-time factory requires significant advances
in the factory’s information technology infrastructure. The main challenge is to design
a comprehensive IT architecture for advanced manufacturing engineering based on a
flexible solution that integrates all manufacturing environments across the entire prod-
uct lifecycle.
SQL4WL - Optimization of Data Processing in Business Processes
Personnel: M. Vrhovnik, H. Schwarz
Funded by: Universität Stuttgart, IBM (partially)
Duration: 1.1.2006 - 30.4.2010
Enterprises organize their business activities in terms of workflows. Data management
is usually an integral part of these workflows. In this project, we develop concepts for
the optimization of data processing and data management in such data-intensive work-
flows. In particular, we cover the appropriate architecture, optimization rules, control
strategies and internal workflow representations for an optimization framework.
Query-generating Systems
Personnel: H. Schwarz
Duration: 1.1.2007 - 28.2.2010
Many information systems generate queries in order to access data in heterogeneous
data sources. Starting with a detailed study of the broad spectrum of application areas,

13
we investigate implementation and optimization concepts for such query-generating
systems. This covers the main strategies to generate queries as well as new optimization
concepts, e.g., for heterogeneous data management tasks that are connected by com-
plex control flow.
SFB 627 – Project B1 – Homogenized and Virtualized Model Man-
agement
Personnel: A. Brodt, N. Cipriani, M. Großmann, C. Lübbe, C. Stach
Funded by: Deutsche Forschungsgemeinschaft (DFG)
Duration: 1.1.2003 – 31.12.2010
During the second funding period, this project focuses on advanced data management
and processing infrastructure and concepts for the Nexus world model. This includes
NexusDS, a scalable and flexible data stream processing framework, which offers sup-
port for application-specific operators and data, as well as approaches for integrating
query-response, stream- and event-based interaction patterns.
SFB 627 – Project B5 – Dynamic,Time-Referenced Model Data
Personnel: N. Hönle
Duration: 1.1.2007 – 31.12.2010
To support data analysis and prognosis, Nexus requires concepts for managing time-
referenced data. This includes history servers for storing such data, domain-specific
pre-processing algorithms for reducing the data volume of trajectories and support for
interpolation functions.
Content Management as a Service
Personnel: F. Wagner, A. Moosbrugger
Funded by: IBM (partially)
Duration: 1.6.2007 - 31.5.2009
Due to the vast amount of valuable semi-structured and unstructured information
produced, processed and stored today, appropriate Enterprise Content Management
(ECM) is a crucial task for companies. In this joint project with IBM and the Uni-
versity of Hamburg a service-based infrastructure for ECM is developed which can
dynamically adapt to changing workload situations. Another part of the project was
to integrate the prototype into the IBM eDiscovery Manager as a proof of concept.

14
SRC: SimTech - Data Provisioning for Scientiﬁc Workﬂows
Personnel: P. Reimann
Duration: 1.5.2008 – 30.4.2011
As part of the Cluster of Excellence “Simulation Technology” (SimTech) this project
primarily aims at offering a consolidated data management and provisioning abstrac-
tion for scientific workflows, and in particular for simulations. We develop and com-
pare appropriate data (de-)provisioning techniques and define strategies for optimiza-
tion.
SRC: SimTech – A Framework to Capture Digital and PhysicalTest
Cases
Personnel: P. Janowski
Funded by: Daimler AG
Duration: 1.10.2009 – 30.9.2012
In the automotive industry more and more testing processes are accomplished using
simulation technologies. In order to efficiently assure the claimed functionality of
components, systems and overall vehicles, an appropriate checking process has to be
defined, planned and realized. In this joint project with Daimler AG a framework is
developed to capture both digital and physical test cases. By defining unified methods
for data access, computation and storage, this allows to bridge between the digital and
the physical world and vice versa.
Selected Research Projects
SFB 627 – Project B5 – Dynamic,Time-Referenced Model Data
In this project, we focus on research in storage and analysis of time-series data in His-
tory Servers and History Warehouse.
History Servers should be able to store time-series data of different data types
persistently. Intelligent preprocessing of stream data reduces data volume and update
rate, so that it is possible to store histories of stream data in History Servers based on
standard database management systems. However, the preprocessed data gives value
patterns using time-dependent interpolation functions, and the currently used basic
data types as well as the Nexus meta data model are only able to define stationary time-
dependent data values. Therefore we extended our basic data type definition, so that
it is possible to define time-dependent interpolation functions for data types like float
and geographic types, where continuously defined value patterns make sense, e.g., mea-
surement values of temperature sensors or positions of mobile objects. We also imple-

15
mented a History Server prototype based on a DB2 database management system in-
cluding the DB2 Spatial Extender for geographic data types. The query processing of
this History Server uses bounding box information stored together with the interpola-
tion functions for quick data access, and automatically calculates data values using the
interpolation functions.
For the analysis of time series data in the Smart Factory project (SFB 627 Proj-
ect D1), we developed a so called Smart Factory Warehouse. To extract, transform,
and load the source data into the warehouse, NexusDS (SFB 627 Project B1) is used.
While interpreting the data as stream data, the extraction process is synchronous, i.e.,
new source data will be transferred to the warehouse immediately.
In addition, the Nexus platform hides the schema heterogeneity of the data sourc-
es in a transparency layer, because temporal and spatial data like trajectories are pro-
vided in the Augmented World Model, a specialized data model for context-aware ap-
plications.
The Nexus Prepro-
cessing operator (NPC)
for compressing tempo-
ral stream data as position
histories is useful for the
loading of the warehouse
as well. It transforms sin-
gle measurement values of
sensors into interpolated
trajectories and therefore
guarantees, that for every
point in time for which data has to be loaded, there will be a sensor value that can be
loaded into the warehouse. Modeling the uncertainty of compressed trajectory data is
crucial for correct interpretations and analysis.
Content Management as a Service
In today’s businesses it’s getting more and more important to have a highly produc-
tive Enterprise Content Management (ECM) system to cope with an always increas-
ing amount of structured and unstructured data. This data itself must not only be
managed, archived, and retrieved, it’s also desired to make the information available
through search for easier selection of important documents. This is also relevant for
compliance scenarios, where in case of a law suit all necessary documents have to be
given to court in a certain time frame.
Smart Factory
Main Computer
Data
Warehouse
Wrapper
NexusDS
NPC
Sensors
Wrapper
Wrapper
Data
flow of
the ETL
process

16
Another important factor is, that companies nowadays have an increased interest
in efficient and scalable ECM-services. These services would allow companies using
ECM applications in a utility based, cost-effective manner.
In this joint project with IBM Research & Development Germany in Böblingen
and the University of Hamburg a service-based infrastructure for ECM was developed
which can dynamically adapt to changing workload situations. In this joint research
project as sample ECM application an Email Archiving and Discovery solution, like i.
e. IBMs Content Collector, was selected. A prototype was implemented and used to
validate the research results. In the prototype we developed a distributed data model
for this scenario and used known technologies from P2P-systems to create a scalable
system. Furthermore, we integrated an Apache Lucene based search cluster, that man-
ages itself and that is capable to cope with varying load. Another part of the project
was to integrate the prototype into the IBM eDiscovery Manager as a further proof of
concept.

17
Research Focus
Image understanding denotes the multi-sensorial and knowledge based interpretation
of situations, estimations of intentions and predictions of activities of the perceived
objects. This division defines and develops distributed intelligent systems on macro-
and microscopic scale:
▶ Image Understanding
This scientific focus is devoted to the context based recognition of situations and
the prediction of the behaviour of the detected objects. Such a cognitive inter-
pretation of situations is usually distributed and has to be proved on consistency
before it is transferred in to a world model.
▶ Multi-agent systems
Autonomous systems operate under self-control of their internal states, goals and
actions and act usually without human interactions. Such systems are generally
modelled by multiple agents. Representative applications are the development of
autonomous logistic systems, autonomous mobile systems especially autonomous
cars and traffic management by these cars themselves.
▶ Cognitive and evolutionary Robotics
Cognitive robotic systems are provided with an elaborate cognitive architecture
that cooperates closely with a controlling architecture in order to be able to act
fully autonomously. A top-down approach is used to develop autonomous multi
robot systems like for example in the RoboCup scenario. On a microscopic level
evolutionary robots are developed to investigate collective or swarm intelligence.
▶ Nonlinear dynamics and chaos theory
The dynamic of complex technical and natural systems is usually nonlin-
ear and therefore the prediction of the dynamic behaviour of such systems can
only be done if a powerful simulation and investigation tool is available. AnT
2.2 Applied Computer Science -
Image Understanding
Bildverstehen
70569 Stuttgart
Germany
Prof. Dr. rer. nat. habil. Paul Levi
 Paul.Levi@ipvs.uni-stuttgart.de
 +49-711-7816387
 +49-711-7816250

18
(www.Ant4669.de) is such a tool that supports a broad spectrum of system classes
and analysis methods.
Staff
(from left) (front) T. Dipper, B. Eckstein, K. Häussermann, U. Käppeler, A. Koch, P. Levi, U. Gräter,
M. Schanz, S. Kernbach, B. Schießle, D. Sutantyo, E. Meister
(back) B. Schenke, V. Avrutin, D. Di Marco, F. Schlachter, S. Popesku, O. Zweigle, O. Kernbach
(staff not shown on the photo) F. Aichele, H. Rajaie, A. Tamke

19
Research Projects
www.ipvs.uni-stuttgart.de/abteilungen/bv/forschung/projekte
1. RFC-Stuttgart
(formerly known as CoPS - Cooperative Soccer Playing Robots
Stuttgart)
Personnel: O. Zweigle, H. Rajaie, K. Häussermann, U. Käppeler, A. Tamke, A.
Koch, B. Eckstein, M. Schanz
RoboCup is an international joint initiative to promote artificial intelligence (AI), ro-
botics, and related fields. In 2009 the robots of our Team 1.RFC Stuttgart were im-
proved both in hard- and software. Thus an accurate passing system has been devel-
oped which was able to adapt the passing behavior during the game. Furthermore we
improved the learning algorithms to adapt the tactics and strategic behavior in real-
time as well was the motor controller to improve the driving behavior has been opti-
mized.
RoboEarth - Connecting Robots Worldwide
Personnel: A. Koch, K. Häussermann, B. Schießle, D. DiMarco
Funded by: European Community (EC)
Duration: 1.12.2009 - 30.11.2013
RoboEarth is a world-wide platform which enables robots to exchange position and
map information as well as task-related, hardware-independent action recipes. In 2009
we developed first approaches to create the general architecture. Furthermore we dis-
cussed and evaluated some definitions and agreements of the centralized and platform
independent web-based database system, which will be able to manage different ro-
botic relevant data.
SFB 627 - Project E3 - Distributed Situation Recognition Based on
Evaluated Context Information
Personnel: K. Häussermann
Duration: 1.1.2007 - 31.12.2010
The objectives of sub-project E3 is to develop a distributed application specific situa-
tion recognition based on evaluated context information. In the year 2009 a new ap-
proach has been developed to calculate and adapt the situation templates, based on
the quality of the context sources. Furthermore a method to detect and solve inconsis-
tencies using a so called meta-template has been developed and evaluated accordingly.

20
SFB 627 – Project C3 – Sensor-Supported, Context-Based Consis-
tency Analysis
Personnel: U. Käppeler
Duration: 1.1.2003 – 31.12.2010
One of the main objectives of the subproject C3 is to develop concepts for the evalu-
ation of the quality of sensor data, in order to integrate indistinct measurements and
inaccurate results into distributed environment models. In the year 2009 a method to
derive higher level context or situation information combined with probabilities from
inaccurate sensor data using a multinomial logistic regression has been developed and
evaluated successfully.
Replicator – Robotic Evolutionary Self-Programming and Self-As-
sembling Organisms
Personnel: S. Kernbach, E. Meister, S. Popesku
Duration: 1.3.2008 - 28.2.2013
This project as well as the project SYMBRION are both EC-funded projects. There
are big synergies between both projects however some essential points differ. In REP-
LICATOR the main goal is to build a sensor- and communication-rich platform that
are capable of self-aggregation into large-scale artificial organisms. Flexible locomo-
tion and adaptation are the main focuses and should be achieved autonomously with-
out human intervention. A big challenge is to develop a highly adaptive hardware in
cooperation with highly adaptive software.
Symbrion – Symbiotic Evolutionary Robot Organisms
Personnel: S. Kernbach, O. Kernbach, F. Schlachter
Duration: 1.2.2008 - 31.1.2013
The main focus of this project is to investigate novel bio-inspired approaches for sym-
biotic multi-robot organisms. Principles of evolution and adaptation in large-scale ar-
tificial organisms play the main role of the research focus. Based on experiences and
knowledge from former projects such as I-SWARM or MiCRON, which both dealt
with emergent swarm behavior, as a next step aggregation into organisms that are phys-
ically connected have been developed. As a result at the end of this project, it is ex-
pected that such artificial organisms will be capable of self-configuration, self-healing,
self-adaptation or self-protection and hence human intervention should be avoided.
Such systems can be used for rescue tasks in hazardous environments where presence

21
of humans either not possible or perilous. The SYMBRION project involves several
beneficiaries from the REPLICATOR projects and vice versa.
ANGELS – Anguilliform Robot with Electric Sense
Personnel: S. Kernbach, D. Sutantyo
Duration: 1.2.2009 - 31.1.2012
The aim of the ANGELS project is to design and build a prototype of a reconfigu-
rable swimming robot able to split into smaller agents (and vice-versa). Compared to
the present existing solutions for Autonomous Underwater Vehicles (AUV) explora-
tion, ANGELS first exploits the advantages of using a bio-inspired sense that has not
been extensively explored by robotics until now: the electric sense. On this point, a
new promising sensor technology is introduced, that can be applied in environmental
exploration and inspection where vision cannot be used. Moreover, in order to inves-
tigate the underwater collective behavior, a test platform robot is also prepared. The
ANGELS test platform uses a small size underwater swimming toy as the mechanical
platform and blue light sensor for communication among robots.
www.ipvs.uni-stuttgart.de/abteilungen/bv/forschung/projekte
1. RFC-Stuttgart (formerly known as CoPS - Cooperative Soccer
Playing Robots Stuttgart)
RoboCup is an international joint initiative to promote artificial intelligence (AI), ro-
botics, and related fields. It is an attempt to boost robotic research and artificial intel-
ligence by providing a standard problem where a wide range of technologies can be in-
tegrated and examined.
The 1. RFC-Stuttgart (for-
merly known as the CoPS
team) is a middle size team
of the Department of Image
Understanding of the IPVS.
In 2009 the new redesigned
robot hardware with a four
wheeled omnidirectional
drive was enhanced in sev-
eral parts of hardware and
software. The hardware has

22
been enhanced by new constructed wheels and by the development of a new digital
compass to improve the localization of the robot. Furthermore the adjustable electro-
magnetic kicker device was boosted to allow stronger ball shoots.
One of the major software topics in 2009 were the development of an accurate
passing system which is able to adapt the passing behavior during the game, the im-
provement of learning algorithms to adapt the tactics and strategic behavior in real-
time and also the optimization of the motor controller to improve the driving behav-
ior. Furthermore to analyze a game in real-time an autonomous camera-agent was
developed, which is driven only by communicated robot data in combination of a
highly sophisticated 3D Visualization.
The team won the German Open 2009 in Hanover and even during the world
championship of Robocup in Graz (Austria), the 1.RFC-Stuttgart became new world
champion 2009 in the the middle-size league. Furthermore the team achieved the 2nd
place at the so-called “technical challenge” and a further 1st place at the “scientific
challenge”.
RoboEarth - Connecting Robots Worldwide
RoboEarth is a world-wide platform which robots can use to exchange position and
mapinformationas well as task-related,hardware-independent actionrecipes. Thiswill
enable manufacturers worldwide to break down their costs and efforts for reproducing
software algorithms for robot behavior over and over again. The RoboEarth frame-
work can store all relevant data from algorithms to complex behavior descriptions that
allows robots to act autonomously in an unknown, unspecified environment. In 2009
some literature surveys and first approaches have been developed to create the general
architecture of the so called recognition-/labeling-component, which will serve as the
basic interface between the robot platforms and RoboEarth. Furthermore first defi-
nitions and agreements of
the architecture of the cen-
tralized and platform in-
dependent web-based da-
tabase system, which will
be able to manage differ-
ent robotic relevant data,
has been discussed and
evaluated.

23
Research Focus
The research activities of the department Parallel Systems involve the analysis and per-
formance optimization of hardware/software systems with the focus on the interac-
tion of the different levels of the system design. This includes the physical design level
of the hardware platform, the architectural level of the devices and the algorithmic
level of the application. Prototypes are built to demonstrate the efficiency of the ap-
proach. In order to obtain very compact high performance systems the resulting hard-
ware/software systems are domain specific with a focus on the following research areas:
▶ Parallel Real-Time Image and Video Processing
In many industrial scenarios like the monitoring of manufacturing processes, the
production process is controlled using image and video processing. Today’s im-
age sensors enable resolutions well above one megapixel with frame rates exceed-
ing several hundred fps which are well suited for monitoring even fast processes.
While the raw video material can be obtained easily, its analysis using CPU-based
hardware and algorithms implemented in software is very time-consuming due
to the huge amount of data involved. In the case of several hundred of high-reso-
lution images per second real-time processing can be achieved by reconfigurable
hardware in conjunction with highly-parallelized algorithms. In order to apply
this to process control, the latency of the algorithms mapped on the specific hard-
ware platform is an important issue.
▶ Hardware Accelerators for Scientific Computing
In order to reduce the computation time in scientific computing, typically PC
clusters or supercomputers are used. Usually, these computing resources have to
be shared with other users due to cost reasons such that a certain waiting time
or latency has to be accepted by the user which is the contrary of the intended
acceleration. As an alternative, hardware accelerators based on GPGPUs or re-
configurable hardware (FPGAs) can be used in personal computers such that the
2.3 Parallel Systems
Parallele Systeme
70569 Stuttgart
Germany
Prof. Dr.-Ing. Sven Simon
 Sven.Simon@ipvs.uni-stuttgart.de
 +49-711-7816450
 +49-711-7816250

24
computing resources are available when required. The programming model and
architecture of hardware accelerators differ very much from classical parallel com-
puters such that there is a high demand of research to investigate the achievable
performance in certain application domains.
▶ Hardware Design Techniques for Multi-Gigabit Systems
The computational efficiency of highly parallel hardware architectures requires
multi-gigabit data transfer between different hardware components. Although
system-on-the-chip solutions integrate components with sufficient bandwidth,
the off-chip data transfer bandwidth to external components has become a criti-
cal issue over the years. In order to implement systems with multi-gigabit inter-
connects in the range of 10Gbits/s to 100Gbit/s or above in conjunction with
complex IC devices new design techniques for the integration of heterogeneous
components have to be developed. Particular effort is made to enable accurate and
fast simulation of the physical level of these systems by appropriate measurement
and modeling. Important aspects are signal propagation, power distribution, vari-
ation of manufacturing parameters and signal integrity.
Staff
Staff of the
Department:
(from left)
S. Simon,
M. Shaikh,
L. Rockstroh,
M. Wróblewski,
J. Hillebrand,
T. Bringewat,
U. Gräter,
S. Wahl,
W. Li
(staff not shown on the
photo)
M. Grégoire,
C. Qiu,
Z. Wang

25
Research Projects
www.ipvs.uni-stuttgart.de/abteilungen/pas/forschung/projekte
High Performance Hardware Platforms
Personnel: M. Wróblewski, J. Hillebrand
Funded by: University of Stuttgart
Duration: Unlimited
The goal of the work is the design of high performance hardware platforms based on
reconfigurable devices, general purpose processors and graphics processing units for
specific application domains.
Numerical accuracy analysis in scientiﬁc computing
Personnel: W. Li
Funded by: Simtech, Deutsche Forschungsgemeinschaft (DFG)
Duration: 1.8.2008 – 30.7.2011
Numerical accuracy analysis is a general concern in scientific simulation applications
to reduce the risk of misinterpretation of inaccurate results due to finite precision
computation and rounding error propagation. The research in our project provides an
effective and efficient numerical accuracy analysis method as well as an easy to use nu-
merical accuracy debugging tool.
Enhancement of the Effective I/O-Bandwidth of High Performance
Systems
Personnel: S. Wahl, Z. Wang
The increasing difference between the performance of processor cores and the I/O-
bandwidth leads to limitations which can be overcome by improved effective I/O-
data. This efficiency enhancement is especially applicable in reconfigurable hardware
systems on the bit-level but can also be extended to the general purpose processor do-
main.
A Sensor Data Platform for Monitoring Production Processes
Personnel: L. Rockstroh
Funded by: Graduate School advanced Manufacturing Engineering (GSaME)
Duration: 1.5.2008 – 30.4.2012
The focus of this project is a scalable hardware platform for sensor data processing in a
production environment with emphasis on real time constraints and robustness.

26
A WirelessTransmitter Concept Based on Sigma-Delta-Synthesizers
Funded by: Bundesministerium für Bildung und Forschung (BMBF)
Personnel: K. Merkle – external
Duration: 30.9.2007 – 30.11.2010
In this external research project a circuit concept for a wireless transmitter is investi-
gated which replaces the A/D-converter of a wireless transmitter in order to simplify
the hardware concept.
Acceleration of Algorithms for In-Process MeasurementTechniques
of Nanostructured Surfaces
Personnel: M. Shaikh, M. Grégoire
Duration: 15.6.2009 – 14.6.2011
In this project the algorithmic part of measurement techniques with respect to ac-
celeration is examined which can be applied to nanostructured surfaces. The work is
carried out in collaboration with the Institute of Measurement Techniques of Bremen
University.
Acceleration of Simulations of Electromagnetic Wave Propagation
in Biomedical Applications
Personnel: M. Grégoire, T. Bringewat
Funded by: Baden-Württemberg Stiftung
Duration: 15.6.2009 – 14.6.2011
In collaboration with the “Institut für Lasertechnologien in der Medizin und
Meßtechnik” at the University of Ulm the accelerated simulation of electromagnetic
wave propagation in biological cells are examined based on the classical finite differ-
ence time domain (FDTD) approach and the discrete dipole approximation (DDA)
approach.
Modeling and Analysis of High Speed Multi-Gigabit Interconnects
Personnel: J. Hillebrand
The project aims to derive design and measurement methodologies for board-level in-
terconnects of future multi-gigabit data transmission standards. The focus lies on ex-
perimental investigation, analysis and subsequent modeling of physical effects relevant
to the design of printed circuit boards carrying the signals of these high speed com-
munication standards.

27
A Sensor Data Platform for Monitoring of Production Processes
Thermal spray processes are particle-based manufacturing techniques that deposit
coating material on the surface of work pieces. The feedstock material used for thermal
spraying consists of particles with diameters in between 0.1 and 50 micrometers that
are heated to a molten state and sprayed on the work piece with velocities of up to 900
meters per second. Both, coating quality and the amount of overspray (wasted feed-
stock material) are correlated to particle characteristics such as distribution and veloci-
ties of the particles. Thus measuring the distribution and velocities of the particles is
essential in order to predict coating quality and overspray.
Continuous Particle Image Velocimetry (CPIV) is a method for measuring par-
ticle distribution and velocity. It is based on motion blurring of particles due to the
exposure time of the imaging system and the velocity of the particles. The measure-
ment setup of Continuous Particle Image Velocimetry (CPIV) is the least complex
setup compared to other image-based methods such as Classical Particle Image Ve-
locimetry and Particle Tracking Velocimetry which predestines Continuous PIV for
industrial applications. Current software-based implementations of Continuous PIV
achieve around a few measurement samples per second due to image analysis computa-
tions with complex data dependencies. However, online process monitoring and con-
trol loops require one order of magnitude higher measurement sample rates which can
be achieved by using a highly parallelized and optimized hardware accelerator archi-
tecture implemented on a Field Programmable Gate Array (FPGA).
Implementing a hardware-accelerated version of Continuous PIV requires devel-
oping an efficient algorithm in order to utilize the parallel resources of the FPGA. The
algorithm consists of the following steps:
▶ Step 1: Segmentation of the Image
▶ Step 2: Classification and Particle Velocimetry calculations
(a) original image (b) based on maximum intensity
(c) based on mean intensity (d) combined (b and c)
Segmentation of Continuous PIV
images

28
The first step is to determine for each pixel whether it belongs to the foreground, which
is considered to consist of objects, or to the background of the image. Values of pixels
that are segmented into the foreground are maintained as these values are required for
the classification step while pixels of the background are set to zero. The decision cri-
terion for the segmentation is the intensity of the pixel in dependence to the mean in-
tensity and the largest intensity of neighbouring pixels. This combination enables the
detection of dim particles in noisy environments such as shown in the figure above as
well as the detection of bright particles.
During classification in step 2, each cluster of non-zero pixels is considered as an
object and the extent of these objects is determined by analyzing for each pixel of the
cluster, how it contributes to the length and width of the object.
The figure below illustrates the projections of particle objects. Long cylindrical
objects without significant change of width and intensity are considered particle pro-
jections. Based on start and end position of these projections, length and heading are
determined. Finally, the velocity is calculated by taking the edge length of the image
sensor and the magnification factor into account.
Accelerating the Continuous PIV algorithm is achieved by an efficient hardware-
architecture that is implemented on a FPGA. The architecture consists of two pro-
cessing units that are arranged in a pipeline structure and each of them is related to
one step of the algorithm. After an initialization stage, both units work concurrently
which results in a speedup compared to purely sequential processing. In addition, each
of them is adapted specifically to its task and contains several processing elements that
are arranged in a structure of parallel pipelines.
Numerical accuracy analysis in scientiﬁc computing
The focus of this project is to investigate the impact of a reduced floating point mantis-
sa word length on performance enhancements of floating point algorithms like sparse
matrix solvers. Usually, significant hardware savings can be obtained on hybrid high
performance systems (GPUs, FPGAs) by the reduction of the mantissa word length
for arithmetic operations which can be used for further parallelization and speed up.
(a) single particle (b) overlapped (width)
(c) overlapped (intensity) (d) fractions of a particle
Particle objects occurring
in continuous PIV images

29
However, for any given algorithm, the influence of the word length reduction on the
accuracy of the final result is not obvious. Even worse, it is open whether the classi-
cal double precision mantissa word length is sufficient for the accuracy of widely used
simulation software packages, taken into account that these packages contain millions
of lines of code implemented over years by a large group of persons.
In order to verify the dependency of numerical accuracy on the mantissa word-
length a method based on Discrete Stochastic Arithmetic is used. The finite word
length effects in the arithmetic logic are randomized by changing the behavior (round-
ing, truncation, etc.) for every operation executed, thus yielding each time a different
result R, and therefore a different deviation ε=R-r from the mathematically exact re-
sult r. This enables obtaining of statistically significant results based on only a few re-
peated executions of a particular algorithm.
The method has been shown to work well with several data flow examples (e.g.
matrix-multiplication) as well as some widely used software packages (e.g. LAPACK,
FFTW, R) involving billions of floating point operations. In all these experiments, the
used method showed its effectiveness and reliability. The method has also been applied
to more complex programs like MOLPRO, a quantum chemistry package with several
100.000 lines of source code. One focus of the investigations is the hardware accelera-
tion of the proposed method.
P Random rounding
analysis of the MOLPRO
package. Probability P
(yellow) of obtaining a
particular deviation ε from
mathematically exact result
vs. the value of the deviation
ε. The interpolated curve
(red) approaches the Gauss
distribution (blue)

30
Research Focus
Prof. Marc Alexander Schweitzer, from the Institute for Numerical Simulation at Uni-
versity of Bonn, accepted in early 2010 the position of head of chair and started in
May. Prof. Schweitzer’s research focus has been on mesh-free methods for the solution
of partial differential equations, multilevel and algebraic multigrid methods, as well as
on parallelization and efficient algorithms for simulation problems.
Jun.-Prof. Michael Bader, formerly at the Department of Informatics at Tech-
nische Universität München, started as a SimTech Juniorprofessor in December 2009.
He leads a young researcher’s group on Simulation Software Engineering. His research
focus is on efficient algorithms in numerical simulation, especially on efficient ap-
proaches for parallel adaptive mesh refinement.
Research Projects
www.ipvs.uni-stuttgart.de/abteilungen/sgs/forschung/projekte
SRC: SimTech - Recursively StructuredTriangular andTetrahedral
Grids for Dynamically Adaptive Simulations in DUNE
Personnel: M. Bader, K. Rahnema
Duration: 1.2.2010 – 31.1.2013
The project is embedded in the Cluster of Excellence “Simulation Technology” (Sim-
Tech), and integrated in SimTech’s research network on multi-physics simulation. The
goal is to integrate an existing approach for dynamically adaptive grid generation and
grid based numerical simulation into the problem solving environment DUNE. Dem-
onstrator applications will be chosen from porous media flow (in the context of CO2
sequestration) and oceanic wave propagation (Tsunami simulation).
2.4 Simulation of Large Systems
Simulation großer Systeme
70569 Stuttgart
Germany
Prof. Dr. rer. nat. Marc Alexander Schweitzer
 Marc.Alexander.Schweitzer
@ipvs.uni-stuttgart.de
 +49-711-7816465
 +49-711-7816248

31
Research Focus
A wide spectrum of distributed applications involves a potentially large number of
mobile devices, which might substantially differ in performance and degree of mo-
bility. Often those applications are based on various network technologies, including
infrastructure-based networks, mobile ad hoc networks and sensor networks. Further,
many of those applications exploit context information to dynamically adapt their be-
havior according to the current user situation. The research activities of the distributed
systems group focus on system-level aspects of those applications, in particular meth-
ods for distributed data management and communication are investigated. Currently,
our research addresses the following areas:
▶ Context-aware systems
utilize information about the physical world to adapt automatically to the context
of users. Here, we investigate mechanisms for realizing large-scale distributed con-
text models and mechanisms for supporting mobile context-aware applications.
This research is done within the Collaborative Research Center Nexus (SFB 627).
▶ Adaptive Communication Systems
Distributed applications deployed on large scale need to cope with a high degree
of network dynamicity because of changes to the set of participating processes,
failures or mobility of processes. The focus of our research is to support robust
and reliable applications by devising communications abstractions with the abil-
ity to dynamically adapt according to the availability of resources.
▶ Self-organizing software systems
A self-organizing software system is able to adapt its structure to changing envi-
ronmental conditions to optimize its operation. This is especially important for
systems that do not allow manual maintenance due to their scale and dynamics.
Here, we investigate new mechanisms and concepts for the self-organization in
such systems.
2.5 Distributed Systems
Verteilte Systeme
70569 Stuttgart
Germany
Prof. Dr. rer. nat. Dr. h.c. Kurt Rothermel
 Kurt.Rothermel@ipvs.uni-stuttgart.de
 +49-711-7816434
 +49-711-7816424

32
▶ Sensor Networks
Wireless sensor networks have been proposed as a way to unobtrusively gather
real-world data. Typically, they consist of small networked devices equipped with
sensors. The focus of our research lies on the software architecture and algorithms
needed to deal with the increasing complexity of those systems.
▶ Grid Computing
is a discipline of high performance computing which deals with federating re-
sources that belong to different administrative domains and are often globally dis-
tributed. In our research we investigate on how to employ Peer-to-Peer principles
to cope with the high degree of volatility and heterogeneity of resources pooled
in Grid environments.
In the year 2009, the Distributed Systems group together with the Collaborative Re-
search Center Nexus organized the First International Workshop on Quality of Con-
text (QuaCon 2009). QuaCon 2009 was the first scientific event that specifically fo-
cused on the different aspects of quality of context information. It aimed to bring
together researchers from various fields to discuss different aspects of context quality
and to make a consolidated contribution towards an integrated way of treating this
topic. The workshop was well received by more than 100 researches from Europe, Asia,
and USA.
Moreover, members of the Distributed Systems group presented their work at
various prestigious scientific events, in particular, Eighth IEEE International Confer-
ence on Pervasive Computing and Communications (PerCom 2010), Sixth Interna-
tional Conference on Mobile and Ubiquitous Systems (MobiQuitous 2009), and In-
ternational Conference on Mobile Data Management (MDM 2009).

33
Staff
Research Projects
www.ipvs.uni-stuttgart.de/abteilungen/vs/forschung/projekte
ALLOW - Adaptable Pervasive Flows
Personnel: K. Herrmann, H. Wolf, C. Hiesinger, S. Föll
Duration: 1.2.2008 – 31.1.2011
ALLOW is a European research project funded under the 7th Framework Programme.
Its objective is to develop a new programming paradigm for human-oriented adaptable
pervasive applications. The core concepts of the project are Adaptable Pervasive Flows
(APFs). APFs are employed for modeling the activities of the users involved in a per-
(from front) (left row) A. Grau, H. Weinschrott, W. Blochinger, K. Rothermel, S. Schuhmann, A. Benzing,
B. Koldehofe, L. Geiger, M. Wernke, Ch. Hiesinger
(middle row) G. Koch, B. Ottenwälder, M. Brodbeck, S. Rizou, A. Roesler, S. Föll, B. Schilling, M. Völz
(right row) A. Tariq, B. Hameed, P. Skvorzov, S. Schnitzer, D. Fischer, H. Wolf, K. Herrmann, F. Dürr
(staff not shown on the photo) R. Lange, D. Philipp, P. Schlegel, S. Schulz

34
vasive application. This model is then used for adapting the environment to the user
and his current needs.
Parallel SAT Solving Methods
Personnel: W. Blochinger, S. Schulz
Duration: 1.7.2006 - 30.6.2010
SAT solving checks whether there exists a satisfying variable assignment for a given
Boolean formula. It has become an essential tool in many application domains, e.g.,
electronic design automation, software verification, or configuration of complex prod-
ucts. The objective of this project is to significantly speed up SAT solving methods by
parallel processing, extending the range of current and paving the way for future appli-
cations. In the second funding period, we focus on methods and systems for exploiting
the distributed, highly volatile, and heterogeneous resources of Peer-to-Peer Desktop
Grids for parallel SAT solving.
Global Sensor Grid
Personnel: B. Koldehofe, A. Benzing
Funded by: Cluster of Excellence Simulation Technology
Duration: 1.5.2008 - 30.4.2011
The goal of the project is to provide gapless real-time information to simulations on a
global scale. To fill the gaps in existing sensor network deployments, so-called diagnos-
tic simulations are distributed in the Global Sensor Grid. Flexible access to sensor data
is provided by supporting moving queries that can be dynamically relocated to follow
the observed phenomenon. The project is part of and funded by the Cluster of Excel-
lence Simulation Technology.
SpoVNet 2.0
Personnel: B. Koldehofe, G. Koch, A. Tariq
Funded by: Baden-Württemberg Stiftung
Duration: 1.1.2010 - 31.12.2011
SpoVNet 2.0 is a transfer project for the precursory SpoVNet project where a method-
ology and an infrastructure were developed to deploy application-oriented services in
heterogeneous communication networks in a spontaneous and adaptive way. In addi-
tion, SpoVNet 2.0 investigates the modularization and intelligent composition of the
services and their integration into the Internet of Things and adaptation to network
virtualization.

35
Reliability in Distributed Complex Event Processing
Personnel: M. Völz
Duration: 15.1.2009 - 31.12.2010
While distributed Complex Event Processing systems can meet the scalability require-
ments of modern applications, any requirements concerning reliability remain unad-
dressed so far. This project aims to provide distributed CEP systems with the needed
reliability, e.g. by masking failures through replication or reactive failure compensa-
tion. Furthermore, performance aspects such as efficient placement of replicated com-
ponents are considered.
DHEP – Distributed Heterogeneous Event Processing
Personnel: B. Koldehofe, B. Schilling
Funded by: IBM
Duration: 1.3.2008 – 28.2.2010
The project “Distributed Heterogeneous Event Processing” explores new ways in order
to increase the scalability and availability of existing complex event processing tech-
nology by establishing a distributed event correlation network. Therefore, a frame-
work is developed that comprises heterogeneous correlation engines to enable efficient
event processing. The focus lies in providing extensive functionality by the introduc-
tion of a highly expressive description language and adaptive rule placement strategies
that account for the heterogeneity of the underlying network.
Complex Query Optimization in P2P Overlay Networks
Personnel: F. A. Memon
This project deals with the optimization of complex queries in P2P overlay networks,
with main focus on multi-attribute range queries. Given a set of queries that have been
previously monitored in a P2P overlay network, the goal is to optimize the search in-
dex for these queries. Such an optimization leads to significant improvement in overall
system performance in terms of scalability.
UP2P - Uniform Peer-to-Peer Protocols for Heterogenous Networks
Personnel: B. Koldehofe
Funded by: Deutscher Akademischer Austauschdienst (DAAD)
Duration: 1.1.2009 - 31.12.2010
The project on uniform peer-to-peer computing (UP2P) is a research cooperation be-
tween the IPVS and the University of Lisbon to establish and intensify cooperation

36
between young researchers and Master Students in Germany and Portugal. The goal is
to identify overlay structures that can be deployed at low cost over heterogeneous net-
works, for instance over ad-hoc and infrastructure networks.
AWARE - Platform for Autonomous self-deploying and operation of
Wireless sensor-actuator networks cooperating with AeRial objects
Personnel: K. Herrmann, M. Gonzalo
Duration: 5.6.2006 - 4.6.2009
AWARE was a European research project funded under the 6th Framework Pro-
gramme. Its focus was to develop a platform for enabling the cooperation of autono-
mous aerial vehicles with ground wireless sensor-actuator networks, comprising static
and mobile nodes. Especially in disaster relief scenarios, the results of the AWARE
platform will permit operations in areas that are difficult to access and that do not
provide a communication infrastructure. This is achieved by combining diverse sensor
systems. The distributed system research group was involved in the development of a
respective middleware platform. The project was very successfully completed in Au-
gust 2009. The consortium has won the second prize in the “Robotics 2010 Awards”
with the UAV platform that was developed during the project.
NET – Network EmulationTestbed
Personnel: K. Herrmann, A. Grau
Duration: 1.4.2001 - 31.8.2011
The NET project is building an emulation testbed for running evaluations of real soft-
ware systems. New concepts of node and time virtualization are used on a computer
cluster to achieve a versatile emulation environment that uses the cluster’s resources
optimally. In conjunction with efficient placement strategies for virtual nodes, the em-
ulation testbed allows for running large-scale network experiments in minimal time.
Optimized Information Discovery
Personnel: F. Dürr, F. Memon
Duration: 1.12.2006 – 30.11.2010
The optimized information discovery (OID) project deals with optimizing the overall
system performance of multi-attribute range (MAR) queries in distributed-hash-table-
based information discovery systems. The goal of this project is to create and dynami-
cally adapt a set of indices over DHTs. The adaptation process is based upon the his-
tory of MAR queries.

37
GSaME – Project E5 – Novel Communication Architecture for the
Smart Factory
Personnel: F. Dürr, B. Hameed
Funded by: Graduate School advanced Manufacturing Engineering (GSaME)
Duration: 1.8.2008 – 31.7.2012
Project E5 investigates consistency issues in product monitoring inside a smart real-
time factory. As a first step, a consistency stack has been developed that conceptualizes
the different consistency issues into separate layers. In addition to this, an algorithm
has been designed to provide probabilistic guarantees about the observation of ob-
jects and the sequences in which these objects are moving through the assembly lines.
Current work is now focused on developing adaptive strategies to dynamically adjust
the probabilities for correct readings of RFID devices depending upon their historical
performance.
SFB 627 – Project A2 – Context-based Communication
Personnel: F. Dürr, L. Geiger
Duration: 1.1.2007 – 31.12.2010
Project A2 researches protocols to efficiently disseminate messages addressed to users
in a certain context instead of directly addressing individuals. Since an efficient dissem-
ination requires up-to-date information about users, we currently focus our investiga-
tions on aggregation methods for user contexts to lower the update load.
SFB 627 – Project B3 – Hybrid Model Management
Personnel: F. Dürr, H. Weinschrott
Duration: 1.1.2003 – 31.12.2010
The Project B3 of the SFB 627 investigates mechanisms to acquire and manage dy-
namic context data in hybrid system structures where infrastructure-based and ad-hoc
networks are integrated. The acquisition of data is based on mobile devices such as
mobile phones and their integrated sensors. These devices opportunistically read envi-
ronmental conditions and communicate the readings.
SFB 627 – Project B5 – Dynamic,Time-referenced Model Data
Personnel: R. Lange
Duration: 1.1.2007 – 31.12.2010
This project focuses on the management and retrieval of highly dynamic context infor-
mation (e.g., moving objects’ trajectories). For this purpose, we researched protocols

38
for the real-time transmission of trajectory data from moving objects to remote data-
bases optimizing communication costs and storage consumption [LWG+09,LDR10].
Moreover, we proposed a scalable tree-based structure for indexing context providers
by means of formal descriptions of their partial world models [LDR09].
SFB 627 – Project E3 – Distributed Situation Recognition
Personnel: F. Dürr, S. Rizou
Duration: 1.1.2007 – 31.12.2010
Project E3 investigates methods for the realization of a distributed situation recogni-
tion system. As a first step, we introduced a system architecture that enables distribut-
ed context reasoning in large-scale scenarios [RHD+10]. Moreover, we developed an
algorithm for the efficient distribution of the reasoning operators that minimizes the
overall network load [RD+09]. Currently the research is focused on operator place-
ment strategies that provide Quality of Service guarantees in terms of timeliness to
the users.
SpoVNet 2.0
A basic requirement for the Future Internet is the seamless integration of hetero-
geneous devices and access technologies to support a spontaneous deployment and
adaptive execution of distributed applications. The SpoVNet project addresses these
requirements by an architecture that
supports overlay-based innovative
communication services to adapt to
the dynamics of the underlying phys-
ical network.
At the IPVS, an exemplary
two-layered overlay-based event ser-
vice was developed which supports
adaptation to QoS restrictions and
provides basic security mechanisms.
EONSON is an event notification
service that provides the semantics of
content-based publish/subscribe systems. Service requirements are ensured without
relying on a dedicated broker infrastructure. In addition, Cordies supports the distrib-
uted detection of complex events by providing an expressive correlation description

39
language and self-organizing distribution algorithms for the placement of correlation
detection operators.
In SpoVNet 2.0, three German universities collaborate to transfer the developed
concepts into applications of the Future Internet. To this end, a scheme to modularize
SpoVNet services and a framework for module composition will be developed. Fur-
thermore, the project focuses on connectional extensions to SpoVNet that address up-
coming key technologies for the Future Internet such as the Internet of Things and
Network Virtualization.
SFB 627 – Project E3 – Distributed Situation Recognition
The goal of Project E3 – a joint project of the departments Image Understanding (BV)
and Distributed Systems (VS) of IPVS – is to build a distributed reasoning system.
Our work includes the design of the architecture of the reasoning system, as well as
algorithms both for context reasoning (BV) and for the efficient distribution of the
reasoning process (VS).
The figure below shows the architecture of the distributed reasoning system
[RHD+10]. At runtime, situation detection is initialized by the creation of a logical
plan, which describes the detection of a concrete situation at a certain location or for
a given object by an operator graph. Subsequently the system finds a mapping of the
operators of the logical plan to physical hosts according to the optimization goal of the
operator placement. Finally, the operators are executed by the selected physical hosts
using the NexusDS operator execution environment.
For the efficient distribution of the reasoning operators, we have worked on meth-
ods for optimized operator placement to improve the scalability of the distributed rea-
soning system. In particular, we have designed a novel distributed operator placement
algorithm [RD+09] that tries to find node coordinates in an intermediate continuous
space, called latency space, such that the bandwidth-delay product of inter-operator
data streams is minimized to optimize the induced network load.
Our current research focuses on the extension of the afore-mentioned place-
ment algo-
rithm to
include us-
er-defined
restrictions
in terms of
the com-
munication
latency of
data.
Auxiliary Services
Context Aware Applica on
Request
Response
Plan
Feedback
A
Layer
text
Rea iy
Logical
Planner
Situa on Template
Repository
Execu on Environment
p
fPhysicalP
Feedback
Adapta on
Rea i
Layer
Physical
ni al
Operator
Distributed
Operator
Execu onDistributed
Operator
Context Broker
Resource Model
Logical Plan
loymentof
p
Plan
Operator
Placement
Placement
World Model
Observable
Context
Dep
W rld
Sensor Data Sta c Data Historical Data Situa on Data del
Layer
Archi-
tecture of
Distri-
buted
Reasoning
System

40
Collaborative Research Center 627 NEXUS — Spatial World Models
for Mobile Context-Aware Applications
www.nexus.uni-stuttgart.de
Coordinator: Prof. K. Rothermel
Vice Coordinator: Prof. B. Mitschang, Prof. P. Kühn
Nexus is an interdisciplinary research center in the field of mobile context-aware sys-
tems and applications at the University of Stuttgart, funded by Deutsche Forschun-
gsgemeinschaft (DFG) since the year 2003. The research program of Nexus integrates
a wide range of disciplines, like Computer Science, Geoinformatics, Industrial Manu-
facturing, and Philosophy of Science and Technology. Nexus is headed by Prof. Ro-
thermel from IPVS and consists of 17 research projects including 6 projects from
IPVS (see Section 2.5 for a detailed description of the individual projects). Research
on globally federated context models constitutes the scientific core of Nexus, enabling
our vision of a “World Wide Space” – an open system where partial context models
from different providers are integrated into a global and highly detailed spatial world
model.
An important research focus of the current funding period is quality of context.
As part of this research effort Nexus organized the First International Workshop on
Quality of Context (QuaCon 2009), which aimed to bring together researchers from
various fields to discuss the different facets of context quality and to make a consolidat-
ed contribution towards an integrated way of treating them. The workshop attracted
The Nexus World
Model connects
the physical and
the virtual world
2.6 Joint Activities
Physical World
Spatial World Model
Digitale Information Spaces

41
about 100 researches from Europe, Asia, and from USA. This great success also under-
lines the importance of research on context quality to the community.
Other major scientific meetings of the Nexus project organized with support of
IPVS members in the year 2009 dealt with the industrial application of context-aware
systems (“Industriekolloquium”) and the management and analysis of context infor-
mation (Workshop “Verwaltung, Analyse und Bereitstellung kontextbasierter Infor-
mationen”). Both were attended by about 80 persons from academia and industry.
As in previous years, Nexus researches from the IPVS present their work on nu-
merous conferences, symposia and workshops. Among others, prestigious scientific
events such as the International Conference on Mobile Data Management (MDM
2009) and the IEEE International Conference on Pervasive Computing and Com-
munications (PerCom 2009) can be addressed. Moreover, Prof. Rothermel presented
a keynote on “Large-scale Context Management” at PerCom 2010. Also prototype
systems putting our research results into practice were presented at major events like
PerCom 2009 and 2010.
Moreover, research results of Nexus were presented at several public events at the
university. In June 2009 Andreas Brodt presented the NexusWeb demonstrator at the
Science Day (Tag der Wissenschaft). NexusWeb is a location-aware web application
which utilizes the built-in GPS receiver of a mobile device to show a map of the user’s
surroundings augmented with context data from the Nexus federation in a single web
page [BC09].
In February 2010, Lars Geiger, Jonas Palauro, and Ralph Lange offered a work-
shop for preparatory school students on tracking moving objects. The students showed
great interest in the necessary scientific backgrounds for novel mobile or location-
based services. At the same occasion, Andreas Brodt offered a workshop titled “Da-
tabases and Google maps” in which pupils constructed queries on a spatial database
to drive a geo-mashup web application. Both workshops well achieved their goal to
fascinate pupils of the exciting challenges for a new generation of computer science
students.

42
Scope
The group “Infrastructure” is responsible for all tasks referring to the administrative
and technical management of the IPVS. At the time the institute had been founded,
the responsible persons decided to join the administrative as well as technical resources
in one department to create a skilled and efficient support team for the institute.
The department acts today as a service provider for the five operational research
groups of the institute. Project coordination, administrative management and infor-
mation technology support are the major areas of responsibility.
Project Coordination: The main focus comprises the support of the acquisition and
management of national as well as European collaborative research activities in basic
and applied research for all the other departments. Among others the administrative
and financial management of the Collaborative Research Center 627 – Nexus is set-
tled here since 2003. Since 2008 the administrative coordination of three collaborative
projects funded by the European Community has been added. These tasks as well as
the management of the department itself, the organization of workshops and confer-
ences or the publication of reports are being taken up by the project coordinator.
Administrative Management: The management joins the resource management
(personnel and finances) of the institute’s budget as well all third party collaborative
projects. In particular the daily financial accounting transactions, purchases, personnel
administration, travel arrangements and much more are performed.
Information Technology Support: The IT support (Computer Lab) is responsible
for the daily operation and maintenance, as well as planning and purchasing the central
IT resources of the institute. Several hundreds of computer systems ranging from lap-
tops and PCs up to compute servers and clusters are managed by the Computer Lab.
2.7 Infrastructure
Infrastruktur
70569 Stuttgart
Germany
Dipl.-Inf. Michael Matthiesen
 Michael.Matthiesen@ipvs.uni-stuttgart.de
 +49-711-7816414
 +49-711-7816248

43
The local network and the connection to the faculty-wide net is also being adminis-
tered. Additionally all the required basic software services in a heterogeneous OS en-
vironment are provided.
Staff
Staff of the
Department:
(from left)
(front)
H. Kniehl,
S. Palmer,
S. Thielmann,
M. Rasch,
A. Schaupp
(back)
C. Reissner,
R. Aumüller,
F. Fabian,
B. Schusser,
M. Matthiesen

44
Teaching
Number of DiplomaThesis: 30
Number of MasterThesis: 11
Number of StudentThesis: 13
Number of Student Projectsts: 4

4545
3 Teaching
www.ipvs.uni-stuttgart.de/lehre
The Institute participates in the following study programs:
Diploma in Computer Science: Computer science generally uses formal (mathe-
matical) and engineering-based methods to solve problems. Both software and hard-
ware systems form the focus of the program, both as research objectives and as tools.
Diploma in Software Engineering: This course is oriented towards the everyday
tasks of software developers and managers in industry. Students learn how to specify,
design, and implement software systems, and are introduced to software project man-
agement and quality assurance. Large projects and team-work are essential aspects.
This course, which is unique in Germany, has been awarded a national prize and been
given an excellent rating by an international evaluation team.
Msc. in InformationTechnology: This is an international master program. Informa-
tion Technology stands for enabling technologies like hard- and software design, elec-
tronic and photonic components, computer and communication systems, communi-
cation networks, multimedia, traffic and process control systems. It further stands for
innovative solutions in areas like mobile/fixed communication, Internet, automotive
and industrial automation.
Bsc. in Information Systems: The ubiquity of information systems and commu-
nications systems in today‘s economy makes it necessary to have a profound under-
standing of the interrelationships between business processes, organisational design
and the use of IT. Teaching methodological expertise and professional competence for
analysis, design and implementation of IT solutions is the primary aim of the program.
Basics in mathematics and statistics are part of the program as well as classes in social
sciences, economics, business administration and computer science.
Msc. in Information Systems: The program offers a high number of special courses
and course clusters in social sciences, economics, business administration and comput-
er science to deepen the topics introduced in the Bsc. program.

46
Dissertations
and
Postdoctoral
Qualiﬁcations

47
4 Dissertations and
Postdoctoral Qualiﬁcations
Dissertations
Buchheim,Thorsten
Kooperative Verhaltensmodellierung für adaptive Multiagentensys-
teme
Supervisors: Prof. Levi, Prof. Roller
The focus of the present dissertation is the modelling of cooperative behavior for
groups of autonomous robots in dynamic environments. This is done by a formal-
ism to describe interactions in a multi robot system in form of so called special-
ized interaction networks which define the chronological sequence of distribut-
ed tasks in a group of autonomous robots. Furthermore, this work examines the
application of reinforcement learning to optimize the performance of the group
behavior. This is achieved on one hand by improving a single agent’s behavior,
on the other hand by improving the assignment of tasks to the individual robot
agents. The latter is achieved by a cost based approach which uses invididual cost
estimates of the agents for performing certain tasks to calculate a cost minimal
assignment of tasks to robots. By approximating these cost estimates through a
reinforcement learning approach the group performance is gradually improved.
Dudkowski, Dominique
Fundamental storage mechanisms for location based services in mo-
bile ad hoc networks
Supervisors: Prof. Rothermel, Prof. Marrón
The proliferation of mobile wireless communication technology has reached a
considerable magnitude. As of 2009, a large fraction of the people in most in-
dustrial and emerging nations is equipped with mobile phones and other types
of portable devices. Supported by trends in miniaturization and price decline of
electronic components, devices become enhanced with localization technology,
which delivers, via the Global Positioning System, the geographic position to the
user. The combination of both trends enables location-based services, bringing in-
formation and services to users based on their whereabouts in the physical world,
for instance, in the form of navigation systems, city information systems, and

48
friend locators. A growing number of wireless communication technologies, such
as Wireless Local Area Networks, Bluetooth, and ZigBee, enable mobile devices
to communicate in a purely peer-to-peer fashion, thereby forming mobile ad-hoc
networks. Together with localization technology, these communication tech-
nologies make it feasible, in principle, to implement distributed locationbased
services without relying on any support by infrastructure components. However,
the specific characteristics of mobile ad-hoc networks, especially the significant
mobility of user devices and the highly dynamic topology of the network, make
the implementation of locationbased services extremely challenging. Current re-
search does not provide an adequate answer to how such services can be support-
ed. Efficient, robust, and scalable fundamental mechanisms that allow for generic
and accurate services are lacking.
This dissertation presents a solution to the fundamental support of location-
based services in mobile ad-hoc networks. A conceptual framework is outlined
that implements mechanisms on the levels of routing, data storage, location up-
dating, and query processing to support and demonstrate the feasibility of loca-
tion-based services in mobile ad-hoc networks. The first contribution is the con-
cept of location-centric storage and the implementation of robust routing and
data storage mechanisms in accordance with this concept. This part of the frame-
work provides a solution to the problems of data storage that stem from device
mobility and dynamic network topology. The second contribution is a compre-
hensive set of algorithms for location updating and the processing of spatial que-
ries, such as nearest neighbor queries. To address more realistic location-based
application scenarios, we consider the inaccuracy of position information of ob-
jects in the physical world in these algorithms. Extensive analytical and numerical
analyses show that the proposed framework of algorithms possesses the necessary
performance characteristics to allow the deployment of location-based services in
purely infrastructureless networks. A corollary from these results is that currently
feasible location-based services in infrastructure-based networks may be extended
to the infrastructureless case, opening up new business opportunities for service
providers.
Handte, Marcus
System support for adaptive pervasive applications
Supervisors: Prof. Rothermel, Prof. Becker
Driven by the ongoing miniaturization of computer technology as well as the pro-
liferation of wireless communication technology, Pervasive Computing envisions
seamless and distraction-free task support by distributed applications that are ex-

49
ecuted on computers embedded in everyday objects. As such, this vision is equally
appealing to the computer industry and the user. Induced by various factors such
as invisible integration, user mobility and computer failures, the resulting com-
puter systems are heterogeneous, highly dynamic and evolving. As a consequence,
applications that are executed in these systems need to adapt continuously to their
ever-changing execution environment. Without further precautions, the need for
adaptation can complicate application development and utilization which hin-
ders the realization of the basic vision.
As solution to this dilemma, this dissertation describes the design of system
software for Pervasive Computing that simplifies the development of adaptive ap-
plications. As opposed to shifting the responsibility for adapting an application
to the user or the application developer, the system software introduces a com-
ponent-based application model that can be configured and adapted automati-
cally. To enable automation at the system level, the application developer specifies
the dependencies on components and resources in an abstract manner using con-
tracts. Upon application startup, the system uses the contractual descriptions to
compute and execute valid configurations. At runtime, it detects changes to the
configuration that require adaptation and it reconfigures the application.
To compute valid configurations upon application startup, the dissertation
identifies the requirements for configuration algorithms. Based on an analysis of
the problem complexity, the dissertation classifies possible algorithmic solutions
and it presents an integrated approach for configuration based on a parallel back-
tracking algorithm. Besides from scenario specific modifications, retrofitting the
backtracking algorithm requires a problem mapping from configuration to con-
straint satisfaction which can be computed on-the-fly at runtime. The resulting
approach for configuration is then extended to support the optimization of a cost
function that captures the most relevant cost factors during adaptation. This en-
ables the use of the approach for configuration upon startup and reconfiguration
during runtime adaptation.
As basis for the evaluation of the system software and the algorithm, the dis-
sertation outlines a prototypical implementation. The prototypical implementa-
tion is used for a thorough evaluation of the presented concepts and algorithms
by means of real world measurements and a number of simulations. The evalua-
tion results suggest that the presented system software can indeed simplify the
development of distributed applications that compensate the heterogeneity, dy-
namics and evolution of the underlying system. Furthermore, they indicate that
the algorithm for configuration and the extensions for adaptation provide a suf-
ficiently high performance in typical applications scenarios. Moreover, the results
also suggest that they are preferable over of alternative solutions.

50
To position the presented solution within the space of possible and existing
solutions, the dissertation discusses major representatives of existing systems and
it proposes a classification of the relevant aspects. The relevant aspects are the un-
derlying conceptual model of the system and the distribution of the responsibility
for configuration and adaptation. The classification underlines that in contrast
to other solutions, the presented solution provides a higher degree of automa-
tion without relying on the availability of a powerful computer. Thus, it simpli-
fies the task of the application developer without distracting the user while being
applicable to a broader range of scenarios. After discussing the related approaches
and clarifying similarities and differences, the dissertation concludes with a short
summary and an outlook on future work.
Kraft,Tobias
Optimization of query sequences.
Supervisors: Prof. Mitschang, Prof. Härder
Query optimization is a well-known topic in database research since the 1970s.
This thesis highlights a special area of query optimization that arises from new
trends in the usage of databases. Whereas in the beginning databases were primar-
ily used for transaction-oriented processing of operative data, today databases are
also used to facilitate reporting and analysis on consolidated, historic data. For
the latter, the data is loaded into a large data warehouse and afterwards it is being
analyzed by the use of tools. The tools used to model the flows that extract the
operative data from the source systems, transform these data and load it into the
data warehouse as well as the tools that process the data stored in the data ware-
house often generate sequences of SQL statements that break down a complex
flow or request into a sequence of computational steps. The optimization of this
kind of sequences with respect to runtime is the focus of this thesis. We propose
a heuristic as well as a cost-based approach for this optimization problem. The
cost-based approach is just an enhancement of the heuristic approach. It results
from adding a cost estimation component to the optimizer architecture of the
heuristic approach and by replacing the heuristic control strategy by a control
strategy that considers cost estimates. Both approaches are rule-based approaches
that rewrite a given sequence of SQL statements into a syntactically different but
semantically equivalent sequence of SQL statements. Therefore, we specify a set
of rewrite rules. For cost estimation, we employ the capabilities of the query opti-
mizer of the underlying database management system (DBMS) which is respon-
sible for the execution of the query sequences. To improve the quality of these

51
cost estimates, we support the query optimizer of the underlying DBMS with sta-
tistics that we derive from histogram propagation. For this purpose, we need an
interface that allows to access and manipulate statistics in the underlying DBMS.
Since there exists no standardized interface for this purpose, we define our own
DBMS-independent interface. For the heuristic approach as well as for the cost-
based approach, we provide prototypic implementations in JAVA. Furthermore,
we have implemented the DBMS-independent interface for the three commer-
cial DBMSs IBM DB2, Oracle, and Microsoft SQL Server. We report on the re-
sults of experiments that we conducted with our prototypes and some sample
sequences that we derived by using a commercial tool for online analytical pro-
cessing (OLAP). They show the effectiveness of our optimization approach and
they highlight the optimization potential that lies in rewriting sequences of SQL
statements. Finally, we draw a conclusion and suggest some interesting points for
future research.
Lu, Jing
Constraints and triggers to enhance XML-based data integration sys-
tems
Supervisors: Prof. Mitschang, Prof. Deßloch
XML is becoming one of the main technological integredients of the Internet. It
is now accepted as the standard for information exchange. XML-based data inte-
gration system, which enables sharing and cooperation with legacy data sources,
arises as a more and more important data service provider on the web. These ser-
vices can provide the users with a uniform interface to a multitude of data sources
such as relational databases, XML files, text files, delimited files, Excel files, etc.
Users can thus focus on what they want, rather than think about how to obtain
the answers. Therefore, users do not have to carry on the tedious tasks such as
finding the relevant data sources, interacting with each data source in isolation us-
ing the local interface and combining data from multiple data sources.
Users are always expecting better query performance and data consistency
from the data integration systems. This work proposes an approach to support
constraints and triggers in the XML-based data integration system in order to
optimize queries and to enforce data consistency. Constraints and triggers have
long been recognized to be useful in semantic query optimization and data consis-
tency enforcement in relational databases. This work first gives an approach to use
constraints from the heterogeneous data sources to semantically optimize queries
submitted to the XML-based data integration system. Different constraints from

52
the data sources are first integrated into a uniform constraint model. Then the
constraints in the uniform constraint model are stored in the constraint reposi-
tory. Traditional semantic query optimization techniques in the relational data-
base are analyzed and three of them are reused and applied by the semantic query
optimizer for XML-based data integration system. Among them are detection of
empty results, join elimination and predicate elimination. Performance is ana-
lyzed according to the data source type and the data volume. The semantic query
optimizer works best when the data sources are non-relational, the data volume is
huge and the execution cost is expected to be high.
In order to make the XML-based data integration system fully equipped
with data manipulation capabilities, programming frameworks which support
update at the integration level are being developed. This work discusses how to
realize update in the XML-based data integration system under the Service Data
Objects programming framework. When the user is permitted to submit updates,
it is necessary to guarantee data integrity and enforce active business logics in the
data integration system. This work presents an approach by which active rules in-
cluding integrity constraints are enforced by XQuery triggers. An XQuery trig-
ger model in conformance to XQuery update model proposed by W3C is de-
fined. How to define active rules and integrity constraints by XQuery triggers
is discussed. Triggers and constraints are stored in the trigger repository. The ar-
chitecture supporting XQuery trigger service in the XML-based data integration
system is proposed. Important components including event detection, trigger
scheduling, condition evaluation, action firing and trigger termination are dis-
cussed. The whole XQuery trigger service architecture above a data integration
system is implemented in BEA AquaLogic DataService Platform under the Ser-
vice Data Objects programming framework. Experiments show active rules and
integrity constraints are enforced easily, efficiently and conveniently at the global
level.
Constraints and triggers play an important role in XML-based data integra-
tion systems. Using constraints and triggers in the XML-based data integration
system we can efficiently improve query performance and enforce data consisten-
cy.

5555
5 Publications
Books / Book Chapters
[K10a] Kernbach, S.: Challenges of Pervasive Adaptation: Viewpoint
from Robotics. In “Pervasive Adaptation. Research Agenda for Fu-
ture and Emerging Technologies”. A book by Th. Sc. Community
(Coordinated by A. Ferscha), PerAda, 2010, pp. 26-29, doi
[K10b] Kernbach, S.: Concepts of Symbiotic Robot Organisms. In “Sym-
biotic Multi-Robot Organisms” (Eds.: P. Levi and S. Kernbach),
pp. 5-24, Springer-Verlag, 2010
[K10c] Kernbach, S.: Towards application of collective robotics in indus-
trial environment (chapter in book), In G.G. Rigatos (ed.), Indus-
trial Systems: Modelling, Automation and Adaptive Behaviour,
IGI Global, 2010
[KK10] Kernbach, S.; Kernbach, O.: Structural Self-Organized Control. In
“Symbiotic Multi-Robot Organisms” (Eds.: P. Levi and S. Kern-
bach), pp. 309-328, Springer-Verlag, 2010
[M10a] Meister, E.: Kinematics and Dynamics for Robot Organisms. In
“Symbiotic Multi-Robot Organisms” (Eds.: P. Levi and S. Kern-
bach), pp. 329-339, Springer-Verlag, 2010
[M10b] Meister, E.: Pervasive Adaptation in Robotic Systems. In “Perva-
sive Adaptation. Research Agenda for Future and Emerging Tech-
nologies”. A book by Th. Sc. Community (Coordinated by A. Fer-
scha), PerAda, 2010, pp. 10-11
[MZS+09] Monteiro, R. S.; Zimbrão, G.; de Souza, J. M.; Schwarz, H.;
Mitschang, B.: Exploring Calendar-based Pattern Mining in Data
Streams. In: Nguyen, Tho Manh (ed.): Complex Data Warehous-
ing and Knowledge Discovery for Advanced Retrieval Develop-
ment: Innovative Methods and Applications, IGI Global, 2009

56
[RK10] Redko, V.; Kernbach, S.: Self-learning Behavior of Virus-like Ar-
tificial Organisms. In Symbiotic Multi-Robot Organisms (Eds.: P.
Levi and S. Kernbach), pp. 409-429, Springer-Verlag, 2010
Articles in Journals
[BBS+10] Bader, M.; Böck, C.; Schwaiger, J.; Vigh, C. A.: Dynamically Adap-
tive Simulations with Minimal Memory Requirement - Solving the
Shallow Water Equations Using Sierpinski Curves. SIAM Journal
of Scientific Computing 32(1), p. 212–228, February 2010
[BKZ+09] Benkmann, R.; Käppeler, U.-P.; Zweigle, O.; Lafrenz, R.; Levi, P.:
Resolving Inconsistencies Using Multi-Agent Sensor Systems. In:
Noberto Pires, J. (ed.): Robotica. Vol. 03/09(76), Coimbra: Enge-
book, 2009
[BKR09] Briones, J. A.; Koldehofe, B.; Rothermel, K.: Adaptive Publish/
Subscribe for Wireless Mesh Networks. In: SPINE : 2009
[DB09] Dangelmayr, C.; Blochinger, W.: Aspect-oriented component as-
sembly - a case study in parallel software design. Software: Practice
and Experience, Vol. 39(9), 2009
[GMK+09] Gauger, M.; Marrón, P. J.; Kauker, D.; Rothermel, K.: Low Over-
head Assignment of Symbolic Coordinates in Sensor Networks.
In: Telecommunication Systems, 2009
[HB09] Held, M.; Blochinger, W.: Structured collaborative workflow de-
sign. Future Generation Computer Systems - The International
Journal of Grid Computing: Theory, Methods and Applications,
Vol. 25(6), 2009
[KTK+09] Kernbach, S.; Thenius, R.; Kernbach, O.; Schmickl., T.: Re-Em-
bodiment of Honeybee Aggregation Behavior in Artificial Micro-
Robotic System, In “Adaptive Behavior”, Vol. 17(3), pp. 237-259,
2009

57
[LB09] Lu, J.; Mitschang, B.: Enforcing Data Consistency in Data Integra-
tion Systems by XQuery Trigger Service. In: International Journal
of Web Information Systems. Vol. 5(2), Emerald Group Publishing
Limited, 2009
[SPR09] Schilling, B.; Pletat, U.; Rothermel, K.: Event Correlation in Het-
erogeneous Environments. In: it - Information Technology - Com-
plex Event Processing, 2009
[SBP10] Schulz, S.; Blochinger, W.; Poths, M.: Orbweb - a network sub-
strate for peer-to-peer grid computing based on open standards.
Journal of Grid Computing, Vol. 8(1), 2010
[SBH09] Schulz, S.; Blochinger, W.; Hannak, H.: Capability-aware informa-
tion aggregation in peer-to-peer grids - methods, architecture, and
implementation. Journal of Grid Computing, Vol. 7(2), 2009
Articles in Proceedings
[AKM+09] Alves, S.; Koldehofe, B.; Miranda, H.; Taiani, F.: Design of a Back-
up Network for Catastrophe Scenarios. In: Proceedings of the
ACM International Workshop on Advanced Topics in Mobile
Computing for Emergency Management: Communication and
Computing Platforms (MCEM), 2009
[AES09] Avrutin, V.; Eckstein, B.; Schanz, M.: The bandcount increment
scenario. III. Deformed structures. In: Proceedings of the Royal
Society A. Vol. 465(2101), Royal Society Publishing, 2009
[BS09] Brodt, A.; Stach, C.: Mobile ortsbasierte Browserspiele. In: Gesell-
schaft für Informatik e.V. (Hrsg) (ed.): Tagungsband der 39. GI-
Jahrestagung, 28.9. - 2.10.2009, Universität zu Lübeck

58
[CEB+09] Cipriani, N.; Eissele, M.; Brodt, A.; Großmann, M.; Mitschang, B.:
NexusDS: A Flexible and Extensible Middleware for Distributed
Stream Processing. In: ACM (ed.): IDEAS ‘09: Proceedings of the
2008 International Symposium on Database Engineering & Appli-
cations, 2009
[CL09] Cipriani, N.; Lübbe, C.: Ausnutzung von Restriktionen zur Ver-
besserung des Deployment-Vorgangs des Verteilten Datenstrom-
verarbeitungssystems NexusDS. In: Fischer, Stefan (ed.); Maehle,
Erik (ed.); Reischug, Rüdiger (ed.): Beiträge der 39. Jahrestagung
der Gesellschaft für Informatik e.V. (GI), 2009
[EFH+09] Eberle, H.; Föll, S.; Herrmann, K.; Leymann, F.; Marconi, A.; Un-
ger, T.; Wolf, H.: Enforcement from the Inside: Improving Quality
of Business in Process Management. In: 2009 IEEE International
Conference on Web Services (ICWS 2009), 2009
[GDR09] Geiger, L.; Dürr, F.; Rothermel, K.: On Contextcast: A Context-
aware Communication Mechanism. In: IEEE International Con-
ference on Communications, 2009. ICC ‘09., 2009
[GHL+09] Grossmann, M.; Hönle, N.; Lübbe, C.; Weinschrott, H.: An Ab-
stract Processing Model for the Quality of Context Data. In: Pro-
ceedings of the 1st International Workshop on Quality of Con-
text, 2009
[GHR09] Grau, A.; Herrmann, K.; Rothermel, K.: Efficient and Scalable
Network Emulation using Adaptive Virtual Time. In: Proceedings
of the 18th International Conference on Computer Communica-
tions and Networks (ICCCN 2009), 2009
[GSD+09] Geiger, L.; Schertle, R.; Dürr, F.; Rothermel, K.: Temporal Ad-
dressing for Mobile Context-Aware Communication. In: Proceed-
ings of the Sixth Annual International Conference on Mobile and
Ubiquitous Systems: Computing, Networking and Services (Mo-
biQuitous ‘09), Toronto, ON, Canada, July 13-16, 2009

59
[HBW09] Held, M.; Blochinger, W.; Werning, M.: Bioinformatics Work-
flows with Calvin. In 10th International Conference on Web In-
formation Systems Engineering (WISE), number 5802 in LNCS,
2009, Springer-Verlag
[KBE+09] Koch, A.; Berthelot, A.; Eckstein, B.; Zweigle, O.; Häussermann,
K; Käppeler, U.-P.; Tamke, A.; Rajaie, H.; Levi, P.: Advanced data
logging in RoboCup, In Proceedings of the AMS 2009, Autonome
Mobile Systeme, Karlsruhe, Germany, December 3-4, 2009, Infor-
matik Aktuell: 1-8
[KGS+09a] Käppeler, U.-P.; Gerhardt, A.; Schieberle, C.; Wiselka, M.;
Häussermann, K.; Zweigle, O.; Levi, P.: Reliable Situation Recog-
nition based on Noise Levels. In Proceedings of the First Interna-
tional Conference on Disaster Management and Human Health
Risk, New Forest, September 23-25, 2009, volume 110 of WIT
Transactions on the Built Environment, pages 127–137, New For-
est, UK, September 2009
[KGS+09b] Käppeler, U.-P.; Gerhardt, A.; Schieberle, C; Wiselka, M.;
Häussermann, K.; Zweigle, O.; Levi, P.: Reliable situation recogni-
tion based on noise levels. In Proceedings of the First Internation-
al Conference on Disaster Management and Human Health Risk,
2009
[KLM+09] Kernbach, S.; Levi, P.; Meister, E.; Schlachter, F.; Kernbach, O.:
Towards self-adaptation of robot organisms with a high develop-
mental plasticity. In Proc. of the First International Conference
on Adaptive and Self-adaptive Systems and Applications (ADAP-
TIVE-09), Athens/Glyfada, Greece, 2009, pp. 180-187, doi:
10.1109/ComputationWorld.2009.11
[KMS+09] Kernbach, S.; Meister, E.; Scholz, O.; Humza, R.; Liedke, J.; Ricot-
ti, L.; Jemai, J.; Havlik, J.; Liu, W.: Evolutionary Robotics: The
Next-Generation-Platform for On-line and On-board Artificial
Evolution. In Proc. of the IEEE Congress on Evolutionary Com-
putation (IEEE CEC-2009), May 18-21, Trondheim, Norway,
2009, pp. 1079-1086, doi: 10.1109/CEC.2009

60
[KHS+09] Kernbach, S; Hamann, H.; Stradner, J.; Thenius, R.; Schmickl, T.;
Crailsheim, K.; van Rossum, A. C.; Sebag, M.; Bredeche, N.; Yao,
Y.; Baele, G.; Van de Peer, Y.; Timmis, J.; Mokhtar, M.; Tyrrell, A.;
Eiben, A. E.; McKibbin, S. P.; Liu, W.; Winfield, A. F. T.: On adap-
tive self-organization in artificial robot organisms. In Proc. of the
2009 Computation World: Future Computing, Service Computa-
tion, Cognitive, Adaptive, Content, Patterns (ComputationWorld
2009), pp. 33-43, November 15-20, Athens, Greece, 2009. doi:
10.1109/ ComputationWorld.2009.9
[KMM+09] Koldehofe, B.; Mogensen, M.; Monod, M.; Quéma, V.: Heteroge-
neous Gossip. In: Proceedings of the 10th ACM/IFIP/USENIX
International Conference on Middleware, 2009
[LBC+10] Lübbe, C.; Brodt, A.; Cipriani, N.; Sanftmann, H.: NexusVIS: A
Distributed Visualization Toolkit for Mobile Applications (Dem-
onstration). In: Proceedings of the 8th Annual IEEE International
Conference on Pervasive Computing and Communications (Per-
Com ‘10); Mannheim, Germany, March 2010
[LDR10] Lange, R.; Dürr, F.; Rothermel, K.: Efficient Tracking of Moving
Objects using Generic Remote Trajectory Simplification (Demo
Paper). In: Proceedings of the 8th Annual IEEE Internation-
al Conference on Pervasive Computing and Communications
- Workshops (PerCom Workshops ‘10). Mannheim, Germany.
March 2010
[LWG+09] Lange, R.; Weinschrott, H.; Geiger, L.; Blessing, A.; Dürr, F.; Ro-
thermel, K.; Schütze, H.: On a Generic Uncertainty Model for Po-
sition Information. In: Proceedings of the 1st International Work-
shop on Quality of Context (QuaCon ’09), Stuttgart, Germany,
June 2009
[MJH+09] Minguez, J.; Jakob, M.; Heinkel, U.; Mitschang, B.: A SOA-based
approach for the integration of a data propagation system. In:
IRI’09: Proceedings of the 10th IEEE international conference on
Information Reuse & Integration, 2009

Annual Report

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Annual Report

Similar to Annual Report (20)

Annual Report