EUROPEAN ASSOCIATION FOR SIGNAL PROCESSING

POZNAŃ UNIVERSITY OF TECHNOLOGY PTETiS POZNAŃ SECTION

15th European Signal Processing Conference

EUSIPCO 2007

PROCEEDINGS

Edited by: Marek Domański

Ryszard Stasiński Maciej Bartkowiak

September 3-7, 2007 Poznań, Poland

© 2007 EURASIP Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the EURASIP.

The publisher makes no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility or liability for any errors or omissions that may be made. All papers were reproduced exactly as received, on authors’ responsibility.

ISBN 978-83-921340-2-2 Publisher:

PTETiS Poznań, ul. Piotrowo 3a, 31-138 Poznań

The EUSIPCO 2007 Chairman and Organizing Committee takes great pleasure in inviting you

to attend the 15th European Signal Processing Conference EUSIPCO 2007. This year, the conference is organised in a very central location in Europe, in Poznań, Poland. We hope that organisation of EUSIPCO 2007 in Poland will help to build stronger links between all research communities in Europe, also in Central and Eastern Europe.

Conference activities will take place in the Conference Center of Poznań International Fair located in the central area of Poznań. City of Poznań with its population of 570 000 inhabitants (over 700 000 in the agglomeration) is the fifth city of Poland. Moreover, Poznań with its more than 120 000 students of 22 universities and other institutions of higher education is the third academic centre of Poland, and therefore the city is well predestined to host scientific conferences.

We are very pleased that this year the response of the scientific community to the call for conference papers has been quite strong resulting in 764 submissions that came from all 6 continents. The 221 members of Technical Program Committee invited 1208 reviewers who submitted 2200 reviews, i.e. 2.9 reviews per paper in average. For the Final Technical Program, 518 papers authored by 1284 persons have been accepted to 77 oral sessions and 18 poster sessions. The program of the conference also includes 6 outstanding plenary lectures and 2 EURASIP Fellow Inauguration Lectures, and 10 half-day tutorials covering exciting areas of contemporary signal processing.

Acknowledgements and appreciation are due to all contributors who submitted their proposals

for review to EUSIPCO 2007. We also would like to thank the members of the Technical Program Committee, and the reviewers for offering their time in reviewing the submitted papers. We want to thank the plenary speakers, the special session and tutorial organizers, the session chairmen, and the participants for their efforts towards a successful scientific event. Last but not least, we would like to express our deepest appreciation to Organising Committee and all those who helped in the organization of the Conference.

On behalf of Organising Committee, we are very much looking forward to welcoming you at EUSIPCO 2007 to be held in Poznań in September 2007.

Marek Domański EUSIPCO 2007 Chairman

The conference is organised under auspices of

Ryszard Grobelny Major of Poznań,

and Adam Hamrol Rector of Poznań University of Technology

Organising institutions

EURASIP – European Association for Signal Processing

Poznań University of Technology, Faculty of Electronics and Telecommunications

PTETiS Polish Society for Theoretical and Applied Electrical Engineering Poznań Section

under auspices of

Signals, Network and Electronic Systems Section of Electronics and Telecommunications Committee of Polish Academy of Sciences

Sponsors We wish to thank the following for their contribution to the success of this conference:

European Office of Aerospace Research and Development, Air Force Office of Scientific Research, United Staes Air Force Research Laboratory <http://www.london.af.mil>

Poznań Supercomputing and Networking Center

The Foundation for Development of Poznań University of Technology

Welcome to Poznań Poznań is a city in halfway between Berlin and Warsaw but older than both two capitals. Numerous monuments remind the long history of the city that was one of the capitals of Poland and the main burial place of the first Polish kings. Busy streets, new building and crowds of students prove that Poznań is an important centre of industry, administration banking, science and education. Poznań - City center

Conference venue The Poznań International Fair, one of the oldest institutions of this kind in Europe, is the biggest centre of international trade in Poland, where trade fairs, exhibitions and shows take place.

Congress Center at Poznań International Fair

Poznań University of Technology is famous for its superb location in Poznań, its outstanding level of education and remarkable history. Originating in the State School of Mechanical Engineering founded in 1919, the school was given the status of a technical university in 1955. Today, with 1200 lecturers and over 20 000 students it is ranked one of the biggest academic centres in Poland. Main Campus

Among nine faculties of the university, there is Faculty of Electronics and Telecommunications that hosts 15th European Signal Processing Conference EUSIPCO 2007. Faculty of Electronics and Telecommunications

www.et.put.poznan.pl

New faculty building

Organising Committee Chairman Marek Domański Poznań University of Technology, Poland

Technical Program Chairmen Ryszard Stasiński Maciej Bartkowiak Poznań University of Technology, Poland

Special Sessions Maciej Niedźwiecki Gdańsk University of Technology, Poland

Tutorials Tomasz Zieliński AGH University of Science and Technology, Poland

Local Arrangements Sławomir Maćkowiak Poznań University of Technology, Poland

Secretariate Agnieszka Jarysz Poznań University of Technology, Poland

Paweł Garstecki Tomasz Grajek Agnieszka Jazdon Damian Karwowski Krzysztof Klimaszewski Zbigniew Korus (computers) Adam Łuczak Jarosław Marek Piotr Rydlichowski Bogusława Sałata (administration) Olgierd Stankiewicz Marta Stępniewska Rafał Wahlich (technical support) Tomasz Żernicki (webmaster) Overseas Liaisons

Andrzej Cichocki Brain Science Institute RIKEN, Japan

Wen Gao Peking University, China

Yo-Sung Ho Gwangju Institute of Science and Technology, Korea

Janusz Konrad Boston University, USA

Masayuki Tanimoto Nagoya University, Japan

Scientific Secretariate Poznań University of Technology Chair of Multimedia Telecommunications and Microelectronics Polanka 3, 60-965 Poznań office@eusipco2007.org Phone: +48 61 6653 900 Fax: +48 61 6653 899

Technical Programme Committee Tyseer Aboulnasr Electrical Engineering School of Information Technology and Engineering

(SITE), University of Ottawa, Canada

Burak Acar Electrical and Electronic Engineering Department, Boğaziçi University, Istanbul, Turkey

Sofiene Affes INRS ÉMT , INRS Énergie, Matériaux et Télécommunications, Canada

Ali Akansu Electrical and Computer Engineering Department, New Jersey Institute of Technology, USA

Paavo Alku Laboratory of Acoustics and Audio Signal Processing, Helsinki University of Technology, Finland

Mustafa Altinkaya Department of Electrical and Electronics Engineering, Zmir Institute of Technology, Turkey

Régine André-Obrecht Structuring, Analyzes and Modeling of documents Video and Audio, IRIT, Université Paul Sabatier, France

Michaël Ansorge ESPLAB, Institute of Microtechnology, Université de Neuchâtel, Switzerland

Ross Arun West Virginia University, USA

Selin Aviyente Michigan State University, USA

Atilla Baskurt Chair of Telecommunication Department, INSA Lyon, France

Ewert Bengtsson Uppsala University, Sweden

Kostas Berberidis Department of Computer Engineering & Informatics, and RACTI R&D, University of Patras, Greece

Fabrizio Berizzi Department of Information Engineering, University of Pisa, Italy

Bhaskar D. Rao Electrical & Computer Engineering, Jacobs School of Engineering UCSD, USA

Shuvra S. Bhattacharyya Department of Electrical and Computer Engineering, University of Maryland, USA

Jan Biemond Information and Communication Theory Group, Mediamatics Department, Delft University of Technology, The Netherlands

Ezio Biglieri University Pompeu Fabra, Barcelona, Spain

Rick Blum Dept. ECE, Lehigh University, Bethlehem, USA

Holger Boche Fraunhofer Institute for Telecommunications, Heinrich-Hertz-Institut, Berlin, Germany

Karlheinz Brandenburg Fraunhofer IDMT & Ilmenau Technical University, Germany

Mónica Fernández Bugallo Department of Electrical Computer Engineering, Stony Brook University, USA

Patrizio Campisi Dipartimento Elettronica Applicata, Università degli Studi “Roma Tre”, Italy

Antonio Cantoni University of Western Australia, Perth, Australia

Alice Caplier HDR, 3GIPSA-lab Grenoble, France

Olivier Cappé Signal Department, TELECOM Paris, France

Jonathon A. Chambers Cardiff School of Engineering, Cardiff University, Wales, UK

Jocelyn Chanussot Signals & Images Laboratory LIS / INPG, Grenoble, France

Luis Chaparro Department of Electrical Engineering, University of Pittsburgh, USA

Vincent Charvillat Ecole Nationale Supérieure d´Electrotechnique, d´Electronique, d´Informatique, d´Hydraulique et des Télécommunications (ENSEEIHT), Toulouse, France

Liang-Gee Chen Graduate Institute of Electronics Engineering, National Taiwan University, Taiwan

Leszek Chmielewski Division of Vision and Measurement Systems (PSWiP), Institute of Fundamental Technological Research of the Polish Academy of Sciences (IPPT PAN), Poland

Pei-Jung Chung School of Engineering and Electronics, Institute for Digital Communications, Joint Research Institute for Signal & Image Processing, The University of Edinburgh, UK

Philippe Ciblat Comelec/ENST, Telecom-Paris, France

Andrzej Cichocki Laboratory for Advanced Brain Signal Processing, Riken Brain Science Institute, Japan

Zygmunt Ciota Department of Microelectronics and Computer Science, Technical University of Łódź, Poland

M. Reha Civanlar Koç University, Rumeli Feneri Yolu, İstanbul, Turkey

Elio Di Claudio INFOCOM Dpt., University of Rome “La Sapienza”, Italy

Pierre Comon Lab. I3S, Algorithmes/Euclide-B, Sophia-Antipolis, France

Colin Cowan Electronic and Electrical Engineering, Loughborough University, Leics, UK

Andrzej Czyżewski Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology, Poland

Paul Dan Cristea "Politehnica" University of Bucharest, Romania

Irek Defée Department of Information Technology, Institute of Signal Processing, Tampere University of Technology, Finland

Marco Diani Department of Information Engineering, University of Pisa, Italy

Andrzej Dobrucki Institute Telecommunications and Acoustics, Wrocław University of Technology, Poland

Simon Doclo Dept. of Electrical Engineering - SCD, Katholieke Universiteit Leuven, Belgium

Kutluyil Doğançay Division of Information Technology, Engineering and the Environment, School of Electrical and Information Engineering, University of South Australia, Australia

Bernadette Dorizetti Département Electronique et physique, Institut National des Télécommunications (INT), Evry, France

Bharat T. Doshi The Johns Hopkins University, Applied Physics Laboratory, USA

Anastasios Doulamis Department of Electrical and Computer Engineering, National Technical University of Athens, Greece

Nikolaos Doulamis Department of Electrical and Computer Engineering, National Technical University of Athens, Greece

Bogdan Dumitrescu International Center for Signal Processing, Tampere University of Technology, Finland

Przemysław Dymarski The Institute of Telecommunications, Warsaw University of Technology, Poland

Touradj Ebrahimi School of Engineering, Signal Processing Institute, Ecole Polytechnique Fédérale de Lausanne, Switzerland

Karen Egiazarian Institute of Signal Processing, Transforms and Spectral Techniques Group, Tampere University of Technology, Finland

Yonina Eldar Department of Electrical Engineering, Technion, Israel Institute of Technology, Haifa, Israel

Nurgun Erdol Department of Electrical Engineering, Florida Atlantic University, USA

Cumhur Erkut Laboratory of Acoustics and, Audio Signal Processing, Helsinki University of Technology, Finland

Luca Fanucci Department of Information Engineering - University of Pisa, Italy

Zanuy Marcos Faúndez Department of Telecommunications and Computers Architecture, Signal Processing Group, Escola Universitària Politécnica de Mataró, Barcelona, Spain

Marqués Ferran Signal Theory and Communications Department, Video Processing Group, Technical University of Catalonia, Spain

André Ferrari Astrophysique, Faculté des Sciences, Université de Nice Sophia Antipolis, France

Inbar Fijalkow ETIS / ENSEA - Univ. Cergy-Pontoise, France

Javier Fonollosa Dept. Signal Theory and Comm., Universitat Politécnica de Catalunya, UPC, Barcelona, Spain

Pascal Frossard School of Engineering, Signal Processing Institute, Ecole Polytechnique Fédérale de Lausanne, Switzerland

Jean-Jacques Fuchs IRISA / INRIA, Campus Universitaire de Beaulieu, Rennes, France

Teddy Furon TEMICS / IRISA, Campus de Beaulieu, Rennes, France.

Nikolaos Galatsanos Department of Computer Science, University of Ioannina, Greece

Krzysztof Gałkowski Institute of Control and Computation Engineering, University of Zielona Góra, Poland

Javier Garcia-Frias Department of Electrical and Computer Engineering, University of Delaware, USA

Ömer Nezih Gerek Department of Electrical Engineering, Anadolu University, Eskisehir, Turkey

Gini Fulvio Department of Information Engineering, University of Pisa, Italy

Maria Sabrina Greco Department of Information Engineering, University of Pisa, Italy

Fredrik Gustafsson Division of Automatic Control, Linköping University, Sweden

Martin Haardt Communications Research Laboratory, Ilmenau University of Technology, Germany

Peter Händel Signal Processing Lab, Royal Institute of Technology, Stockholm, Sweden.

Lars Kai Hansen Informatics and Mathematical Modeling, Technical University of Denmark, Denmark

Eberhard Hänsler Institute of Telecommunications, Technische Universitat Darmstadt, Germany

Alfred Hanssen Department of Physics, University of Tromsø, Norway

K. V. S. Hari Department of Electrical Communication Engineering, Indian Institute of Science, Bangalore, India

Werner Henkel International University Bremen, Germany

Ewa Hermanowicz Faculty of Electronics, Telecommunications and Informatics, Gdańsk University of Technology

Richard Heusdens Department of Mediamatics, Faculty of Information Technology and Systems, Delft University of Technology, The Netherlands

Ulrich Heute Institute for Circuit and System Theory, Faculty of Engineering, Christian-Albrechts-University of Kiel, Germany

Takao Hinamoto Electronic Control Lab., Graduate School of Engineering, Hiroshima University, Japan

Franz Hlawatsch Institute of Communications and Radio-Frequency Engineering, Vienna University of Technology, Austria

James R. Hopgood Institute for Digital Communications, Joint Research Institute for Signal and Image Processing School of Engineering and Electronics, The University of Edinburgh, UK

Brian L. Hughes Department of Electrical and Computer Engineering, NC State University, USA

Ebroul Izquierdo Department of Electronic Engineering, Multimedia and Vision Research Lab, Queen Mary University of London, UK

Andreas Jakobsson Department. of Electrical Engineering, Karlstad University, Sweden

Jiří Jan Faculty of Electrical Engineering and Communication, Brno University of Technology, Czech Republic

Søren Holdt Jensen Department of Electronic Systems, Aalborg University, Denmark

Ian Jermyn Project Ariana (CNRS/INRIA/UNSA), INRIA, Sophia Antipolis, France

Mieczysław Jessa Chair of Telecommunication Systems and Optoelectronics, Poznań University of Technology, Poland

Michael Joham Institute for Circuit Theory and Signal Processing, Technische Universität München, Germany

Philippe Joly Institut de Recherche en Informatique de Toulouse (IRIT), Université Paul Sabatier, Toulouse, France

Christian Jutten Signals & Images Laboratory LIS / INPG, Grenoble, France

Peter Kabal Electrical & Computer Engineering, McGill University, Montreal, Quebec, Canada

Thomas Kaiser Department of Communication Systems, Faculty of Engineering, University of Duisburg-Essen, Germany

Aggelos K. Katsaggelos ECE Department, Northwestern University, USA

Michel Kieffer LSS – CNRS – Supélec, Université Paris-Sud, Gif-sur-Yvette, France

Kevin H. Knuth Department of Physics, College of Arts and Sciences, College of Computing and Information, University at Albany, USA

Ted Kok Dept. of Electrical & Electronic Engineering, Hong Kong University of Science & Technology

Ahmet Kondoz Centre for Communication Systems Research, University of Surrey, UK

Janusz Konrad Department of Electrical and Computer Engineering, Boston University, USA

Bożena Kostek Faculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, Poland

Constantine Kotropoulos Department of Informatics, Computer Vision and Image Processing Group, Aristotle University of Thessaloniki, Greece

Denis Kouamé Département de Physique, UFR Sciences et Techniques - Faculté de Médecine, Université de Tours, France

Gernot Kubin Department of Communications and Wave Propagation, FNW - Fachbereich für Nachrichtentechnik und Wellenausbreitung, Institute of Signal Processing and Speech Communication, Graz University of Technology, Austria

Zbigniew Kulka Institute of Radioelectronics, Warsaw University of Technology, Poland

Vibhor Kumar Laboratory of Computational Engineering, Helsinki University of Technology, Finland

Mikko Kurimo Adaptive Informatics Research Centre, Helsinki University of Technology, Finland

Jean-Louis Lacoume Laboratoire des Images et Signaux, INPG/ENSIEG, Domaine Universitaire, Saint Martin d’Heres, France

Míguel Ángel Lagunas Centre Tecnològic de Telecomunicacions de Catalunya (CTTC), Barcelona, Spain

Pascal Larzabal Laboratoire SATIE, École Normale Supérieure de Cachan, Cachan, France

Lieven De Lathauwer ESAT/SISTA, Katholieke Universiteit Leuven, Leuven-Heverlee, Belgium

Panos Liatsis School of Engineering and Mathematical Sciences, CITY University London, UK

Xingzhao Liu Department of Electronic Engineering, Shanghai Jiaotong University, China

Fabrizio Lombardini Department of Information Engineering, University of Pisa, Italy

Philippe Loubaton Electrical Engineering, Université de Marne la Vallée, France

Marco Luise Department of Information Engineering, University of Pisa, Italy

Rastislav Lukac Epson Canada Ltd., Toronto, Canada

Benoit Macq Communication and Remote Sensing Lab (TELE), Université catholique de Louvain, Belgium

Shoji Makino NTT Communication Science Laboratories, NTT Corporation, Kyoto, Japan

Ryszard Makowski Signal Theory Section, Institute of Telecommunications, Teleinformatics and Acoustics, Wroclaw University of Technology, Poland

Detlev Marpe Image Processing Department, Fraunhofer Institute for Telecommunications, Heinrich Hertz Institute (HHI), Berlin, Germany

Nadine Martin Signal and Image Laboratory, National Center of Scientific Research - CNRS, Grenoble, France

Andrzej Materka Institute of Electronics of the Technical University of Łódź, Poland

Stephen McLaughlin Department of Electronics and Electrical Engineering, Signals and Systems Group, University of Edinburgh, UK

Desmond C. McLernon School of Electronic and Electrical Engineering, Institute of Integrated Information Systems, University of Leeds, UK

Christoph Mecklenbräuker Institute of Communications and Radio-Frequency Engineering, Vienna University of Technology, Austria

William L. Melvin Sensors & Electromagnetic Applications Laboratory, Georgia Institute of Technology, Georgia Tech Research Institute, Atlanta, USA

Gloria Menegaz Dept. of Information Engineering, Faculty of Telecommunications, University of Siena, Italy

Vassilis Mertzios National Technical University of Athens, Greece

Rudolf Mester Institute for Applied Physics, Image and Vision Group, University of Frankfurt, Germany

Renato De Mori LIA CNRS, University of Avignon, Avignon, France

Ralf Müller Department of Electronics & Telecommunications, Norwegian University of Science & Technology, Trondheim, Norway

Asoke Nandi Department of Electrical Engineering and Electronics, University of Liverpool, UK

Patrick A. Naylor Communications and Signal Processing Group, Electrical & Electronic Engineering, Imperial College, London, UK

Arye Nehorai Department of Electrical and Systems Engineering, School of Engineering & Applied Science, Washington University in St. Louis, USA

Alessandro Neri Applied Electronics Department, The Digital Signal Processing, Multimedia, and Optical Communications Group, University of Roma Tre, Italy

Tomas Nordström The Telecommunications Research Center Vienna (FTW), Vienna, Austria

Maciej Ogorzałek Department of Electrical Engineering, AGH University of Science and Technology, Kraków, Poland

Jens-Rainer Ohm Chair for Communications Engineering, Institute for Communications Engineering, Aachen University, Germany

Soontorn Oraintara Electrical Engineering, The University of Texas at Arlington, USA

Stanisław Osowski Institute of Theory of Electrical Engineering, Measurement and Information Systems, Warsaw University of Technology, Poland

Björn Ottersten School of Electrical Engineering, Signal Processing, Royal Institute of Technology (KTH), Stockholm, Sweden

Michel Paindavoine Laboratoire d’Electronique, d’Informatique et d’Image, Universite de Bourgogne, Dijon, France

Zdzisław Papir AGH University of Science and Technology, Kraków, Poland

Thrasyvoulos Pappas EECS Department, McCormick School of Engineering and Applied Science, Northwestern University, USA

Steffen Paul Infineon Technologies AG, Secure Mobile Solutions, Munich, Germany

Mirek Pawlak Department of Electrical and Computer Engineering, University of Manitoba, Winnipeg, Canada

Fernando Pereira Instituto Superior Técnico - Instituto de Telecomunicações, Lisboa, Portugal

Ana I. Pérez-Neira Signal Theory and Communications Department, (TSC), Polytechnical University of Catalonia, UPC, Barcelona, Spain

Jean-Christophe Pesquet Laboratoire D'Informatique de L'institut Gaspard-Monge (UMR 8049 CNRS), Université Marne-La-Vallée, France

Béatrice Pesquet-Popescu Signal and Image Processing Department, Ecole Nationale Supérieure des Télécommunications, Paris, France

Alexander Petrovsky Department of Real-Time Systems, Białystok Technical University, Poland

Marian Piekarski Wrocław University of Technology, Poland

Peter Planinšič University of Maribor Faculty of Electrical Engineering and Computer Science, Slovenia

Moshe Porat Department of Electrical Engineering, Technion - Israel Institute of Technology, Haifa, Israel

William Puech Laboratoire d'Informatique, de Robotique et de Microélectronique de Montpellier, University of Montpellier II, France

Anthony Quinn Dept. of Electronic/Electronical Eng., Trinity College, University of Dublin, Ireland

Rudolf Rabenstein Telecommunications Laboratory, Multimedia Communication and Signal Processing, University Erlangen-Nuremberg, Germany

Giovanni Ramponi DEEI - University of Trieste, Italy

Kamisetty R. Rao Multimedia Processing Lab, The University of Texas at Arlington, USA

Carlo Regazzoni Biophysical and Electronic Engineering Department (DIBE), University of Genoa, Italy

Branimir Reljin Faculty of Electrical Engineering, University of Belgrade, Serbia and Montenegro.

Markku Renfors Institute of Communications Engineering, Tampere University of Technology, Finland

Fabrizio Simone Rovati ST Microelectronics, Advanced System Technology (AST), Agrate Brianza, Italy

Markus Rupp Institute of Communications and RF Engineering, Vienna University of Technology, Austria

Shigeki Sagayama Department of Mathematical Engineering and Information Physics Graduate School of Engineering, The University of Tokyo also Intelligent Information Processing Laboratory, Graduate School of Information Science Japan Advanced Institute of Science and Technology, Japan

Philippe Salembier Image Processing Group, Universitat Politécnica de Catalunya, Barcelona, Spain

Saeid Sanei Centre of Digital Signal Processing, School of Engineering, Cardiff University, UK

Steve Sangwine Department of Electronic Systems Engineering, University of Essex, UK

Tapio Saramäki Institute of Signal Processing, Tampere University of Technology, Finland

Hiroshi Saruwatari Graduate School of Information Science, Nara Institute of Science and Technology, Ikoma-shi, Japan

Peter Schelkens Dept. of Electronics and Information Processing (ETRO), Vrije Universiteit Brussel (VUB), Belgium

Ivan Selesnick Electrical and Computer Engineering, Polytechnic University, Brooklyn, USA

Tapio Seppänen Department of Electrical and Information Engineering, Computer Engineering Laboratory, University of Oulu, Finland

Ling Shao Philips Research Laboratories, Eindhoven, The Netherland

Wiesław Sieńko Department of Electrical Engineering, Gdynia Maritime University, Poland

Pradip Sircar Department of Electrical Engineering, Indian Institute of Technology Kanpur, India

Wan-Chi Siu Department of Electronic and Information Engineering, Centre for Multimedia Signal Processing, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong

Władysław Skarbek Department of Electronics and Information Technology, Warsaw University of Technology, Poland

Aljoscha Smolic Fraunhofer Institute for Telecommunications, Heinrich-Hertz-Institut, Berlin, Germany

Bogdan Smołka Silesian University of Technology, Gliwice, Poland

Yves Sorel INRIA Domaine de Voluceau, Rocquencourt Research Unit, France

John Sorensen Copenhagen University College of Engineering, Ballerup, Denmark

Leif Sörnmo Dept. of Electroscience, Signal Processing Group, Lund University, Sweden

Ljubiša Stanković Dept. of Electrical Engineering, University of Montenegro, Podgorica, Montenegro

Paweł Strumiłło Faculty of Electrical, Electronic, Computer and Control Engineering, Institute of Electronics, Technical University of Łódź, Poland

Wonyong Sung EECS/EE School of Electrical Engineering and Computer Sciences/ Electrical Engineering, Seoul National University, Korea

Ananthram Swami Army Research Lab, Adelphi, USA

Jerzy Szabatin Institute of Electronic Systems, Warsaw University of Technology, Poland

Krzysztof Ślot Institute of Electronics, Technical University of Lodz and Computer Sciences at Academy of Economics and Humanities, Lodz, Poland

Ryszard Tadeusiewicz Institute of Automatics, AGH University of Science and Technology, Kraków, Poland

Jarno M. A. Tanskanen Ragnar Granit Institute, Tampere University of Technology, Finland

Murat Tekalp Department of Electrical and Computer Engineering, Center for Electronic Imaging Systems, Center for Future Health, University of Rochester, New York, USA

Fabian J. Theis Bernstein Center for Computational Neuroscience, MPI for Dynamics and Self-Organisation, Göttingen, Germany

Sergios Theodoridis Dept. of Informatics and Telecommunications, National and Kapodistrian University of Athens, Greece

Jean-Philippe Thiran School of Engineering, Signal Processing Institute, Ecole Polytechnique Fédérale de Lausanne, Switzerland

Massimo Tistarelli Department of Architecture and Planning, Computer Vision Laboratory, Universtity of Sassari, Italy

Jean-Yves Tourneret IRIT / ENSEEIHT / TéSA, Toulouse, France

H. Joel Trussell Department of Electrical and Computer Engineering, North Carolina State University, Raleigh, USA

Stefano Tubaro ll Dipartimento di Elettronica e Informazione (DEI), Politecnico di Milano, Italy

Jan Turan Dept. of Electronics and Multimedia Communications, Faculty of Electrical Engineering and Informatics, Technical University of Kosice, Slovakia

Wolfgang Utschick Associate Institute for Signal Processing, Munich University of Technology (TUM), Germany

Luc Vandendorpe Digital Communications Group, Communications and Remote Sensing Laboratory (TELE), Université catholique de Louvain (UCL), Belgium

Peter Vary Institute of Communication, Systems and Data Processing, RWTH Aachen University, Germany

Saeed Vaseghi School of Engineering and Design, Electronic and computer engineering, Brunel University West London, UK

Gregori Vázquez Department of Signal Theory and Communications, Signal Processing Group, Universitat Politècnica de Catalunya, Spain

Jean-Marc Vesin School of Engineering, Signal Processing Institute, Ecole Polytechnique Fédérale de Lausanne, Switzerland

Pedro Gómez Vilda Dept. de Arquitectura y Tecnología de Sistemas Informáticos, Facultad de Informática, Universidad Politécnica de Madrid, Spain

Sviatoslav Voloshynovskiy Stochastic Information Processing (SIP) Group, CUI - University of Geneva, Switzerland

Zhou Wang Dept. of Electrical Engineering, The University of Texas at Arlington, USA

Lars Wanhammar Department of Electrical Engineering, Linköping University, Sweden

Rodolphe Weber Laboratoire d' Électronique, Signaux, Images, University of Orléans, France

Krzysztof Wesołowski Chair of Wireless Communications, Poznań University of Technology, Poland

Jacek Wojciechowski Institute of Electronics Fundamentals, Warsaw University of Technology, Poland

Mariusz Ziółko Department of Electronics, AGH University of Science and Technology, Kraków, Poland

Abdelhak M. Zoubir Signal Processing Group at the Institute of Telecommunications, Technical University Darmstadt, Germany

Reviewers Anne Margot Aaron Emad Abd-Elrady Fahed Abdallah Yousri Abdeljaoued Karim Abed-Meraim Tyseer Aboulnasr Elias Aboutanios Patrice Abry Nicola Acito Olivier Adam Martine Adda-Decker Marc Adrat Monika Aggarwal Ishfaq Ahmad Robert Aichner Heikki Ailisto Aydin Akan Mohamed Akil Naofal Al-Dhahir Mahmood Al-Khassaweneh Laurent Albera Florence Alberge Antonio Albiol Felix Albu Karsten Alecke Daniel Alfsmann Tatiana Alieva Paavo Alku Luciano Alparone Steve Alty Gonzalo Álvarez Agustín Álvarez-Marquina Pierre-Olivier Amblard Kiarash Amiri Dimitris Ampeliotis Régine André-Obrecht Garzelli Andrea Yiannis Andreopoulos Vicky Andronikou Dimitrios Androutsos Ramesh Annavajjala Michael Ansorge Konstantine Antonakopoulos Gianluca Antonini Marc Antonini Saïd Aouada Fabrizio Argenti Vasileios Argyriou Myriam Ariaudo Sabri Arik Tuukka Arola Yusuf Artan Antonio Artés Rodríguez Xavier Artigas Mohamad Assaad Antonio Assalini Piotr Augustyniak

Florent Autrusseau Federico Avanzini Nastooh Avesta Selin Aviyente Yannis Avrithis Saeed Ayat Baran Aydogan Zdenka Babić Olivier Bachelier Hyeon-Deok Bae Ivan Bajla Artur Bal Robert Baldemair Esfandiar Bandari Dominique Barba Sergio Barbarossa Jean-Pierre Barbot Michel Barlaud Mauro Barni Maciej Bartkowiak Joan Bas Patrick Bas Daniel S. Baum Ilker Bayram Fokko Beekhof Azeddine Beghdadi Christian Behrens Adel Belouchrani Amel Benazza-Benyahia Jacob Benesty Emmanouil Benetos Ewert Bengtsson Mats Bengtsson Messaoud Benidir Jenny Benois-Pineau Hugues Benoit-Cattin Kostas Berberidis Scott Berger Fabrizio Berizzi Jens Berkmann Dorizetti Bernadette Riccardo Bernardini Stefano Berretti Massimo Bertacca Yannick Berthoumieu Laurent Besacier Olivier Besson Shuvra Bhattacharyya Mauro Biagi Robert Bieda Jan Biemond Ezio Biglieri Mehrzad Biguesh Ali Bilgin Tom Bishop Dariusz Bismor Joerg Bitzer

Luca Bixio Laure Blanc-Féraud Kostas Blekas Thomas Blumensath Shannon Blunt Janusz Bobulski Hanna Bogucka BÜlent Bolat Miodrag Bolic Antonio Bonafonte Jean-François Bonastre Pascal Bondon Deva K. Borah Adrian Gheorghe Bors Martin Bossert Nikolaos Boulgouris Emmanuel Boutillon Rémy Boyer Gozde Bozdagi Akar Nikola Bozinovic Javier Bracamonte Torben Brack Eyal Braiman Francesco Branchitta Tomás Brandão Karlheinz Brandenburg Robert Bregović Robert L. Brennan Catarina Brites Jean-Marc Brossier Keith Brown Thomas Brüggen Vittoria Bruni Fritz K. Brunner Marcelo Gomes da Silva Bruno Lorenzo Bruzzone Clemens Buchacher Markus Buck Mónica Fernández Bugallo Andreas Burg Alister G. Burr Tom Bäckström Sebastian Caban Manora Caldera Alfonso Camargo Andrei Câmpeanu Patrizio Campisi Raphaël Canals Çağatay Candan Antonio Cantoni Michael Cantoni Stéphane Canu Olivier Cappé Lorenzo Cappellari Amerigo Capria Marco Carli Michael Casey Marc Castella Andrea Fabio Cattoni Joseph R. Cavallaro

James K. Cavers François Cayre M. Emre Celebi Mehmet Çelik Ali Taylan Cemgil Barbara Cerato Christophe Cerisara Marie Chabert Jonathon A. Chambers Benoît Champagne Wai-Yip Chan Yiu-Tong Chan Nitin Chandrachoodan Camilo Chang Dórea Sumohana Channappayya Giannis Chantas Jocelyn Chanussot Dimitris Charilas Delphine Charlet Werayuth Charoenruengkit Charayaphan Charoensak François Charot Vincent Charvillat Vasilios Chasanis Thierry Chateau Marc Chaumont Chuah Teong Chee Jean-Paul Chemla Fangjiong Chen Haihua Chen Hua-Mei Chen Sheng Chen Tung-Chien Chen Ying-Jui Chen Yu-Jen Chen Antoine Chevreuil Zheru Chi Rachel Chiang Shao-Yi Chien Jakub Chłapiński Leszek Chmielewski Nam Ik Cho Kiyoung Choi Thierry Chonavel Ryszard S. Choraś Patrick Chow Pei-Jung Chung Andrzej Cichocki Tolga Çiloglu Zygmunt Ciota Philippe Ciuciu Thorsten Clevorn Marco Cococcioni Christophe Collet Dinu Coltuc François Combet Patrick-Louis Combettes Pierre Comon Aura Conci Paulo Correia

Ivan Corretjer Roberto Costantini Laura Cottatellucci Bill Cowley Alain Crouzil Sergio Antonio Cruces-Álvarez Holger Crysandt Hans L. Cycon Bogusław Cyganek Andrzej Czyżewski Rozenn Dahyot Milos Dakovic Hai Huyen Dam Armin Dammann Hieu Dang Richard M. Dansereau Adriana Dapena Janeiro Rony Darazi Nandan Das Sarat C. Dass Alberto Dassatti Mike E. Davies Carlos A. Davila Manuel Davy Gerard de Haan Rodrigo Caiado de Lamare Alessia De Rosa Christophe De Vleeschouwer Diemer de Vries Christian Debes David Declercq Irek Defée Olivier Deforges Giovanni Del Galdo Lieven De Lathauwer Marc Delcroix Fabio Dell'Acqua Jean-Pierre Delmas Nathalie Delprat Bernard Delyon Grażyna Demenko Didier Demigny Florence Denis Antonin Descampe Michel Desvignes Riadh Dhaou Elio Di Claudio Marco Diani Francisco Díaz Pérez Kevin Dickson Guido K. E. Dietl Michel Dietrich Nikos Dimitriou Zhiguo Ding Paulo Sergio Ramirez Diniz Ali Mohammad-Djafari Divna Djordjevic Petar M. Djurić Igor Djurovic Zbigniew Długaszewski

Andrzej Dobrucki Simon Doclo Gwenael Doerr Kutluyil Doğançay Marek Domański María Elena Domínguez Rajamani Doraiswami Camilo Chang Dórea Milos Doroslovacki Leo Dorst Arnaud Doucet Matthijs Douze Pier Luigi Dragotti Andrzej Drygajło Frédéric Dufaux Charlotte Dumard Bogdan Dumitrescu Florent Dupont Emrullah Durucan Laurent Duval Wolfgang Eberle Stefan Eberli Karen Egiazarian Elie El Khoury John N. Ellinas Stephen J. Elliott Dan Ellis Peter Engelhardt Chiara Ercole Alper Tunga Erdogan Nurgun Erdol Jan Erkelens Cumhur Erkut Ayşın Ertüzün Iñaki Esnaola Gianpaolo Evangelista Jani Even Magnus Evestedt Joan Fàbregas Seppo Fagerlund Mike Fairhurst Nikos Fakotakis David Falconer Christof Faller Luca Fanucci Jérôme Farinas Marcos Faúndez-Zanuy Ulrich Fecker Michael Felsberg Carles Fernández Prades André Ferrari Paulo J. S. G. Ferreira Anne Ferreol Miguel Ángel Ferrer-Ballester Mehmet Fidan Giancarlo Filligoi Simone Fiori Robert F.H. Fischer Sven Fleck Markus Flierl

Jan Flusser Amalia Foka Federico Fontana Karl Foppe Hassan Foroosh Philippe Forster Steven Fortune Massimo Franceschetti Amir Francos Gunther Fritzer Pascal Frossard Jean Jacques Fuchs Krzysztof Fujarewicz Marco Fumagalli Carson Fung Teddy Furon Zbigniew Gajo Nikolaos Galatsanos Krzysztof Gałkowski Paolo Ettore Gamba Sharon Gannot Javier Garcia-Frias Luis Garrido Nikolaos Gatsis Nikolay D. Gaubitch Tomáš Gedeon Bernd Geiser Guillaume Gelle Deniz Gençağa Fredrik Georgsson Evgeny Gershikov Alex B. Gershman Wolfgang Gerstacker Gilles Gesquière Mohammed Ghanbari Tadesse Ghirmai Ovidiu Ghita Hans Wilhelm Gierlich Fulvio Gini Jean-François Giovanelli Mariusz Głąbowski Dušan Gleich Juan Ignacio Godino-Llorente Hana Godrich Heinz Göckler Norbert Görtz Guy Gogniat Miroslav Goljan Nathan A. Goodman Şükrü Görgülü Hanna Goszczyńska Atanas Gotchev Georgios Goudelis Gerhard Graber Marco Grangetto Maria Sabrina Greco Yves Grenier Mislav Grgic Rémi Gribonval Romulus Grigoras

Stefan Grocholewski Peter Gruber Matthias Gruhne Karen M. Guan Gabriele Guarnieri Anne Guérin-Dugué David Guevorkian Christine Guillemot Şevket Gümüştekin Bilge Günsel. Devanur S. Guru Fredrik Gustafsson Jürgen Götze Michaela Haberler-Weber Emanuël Anco Peter Habets Eberhard Hänsler Dimitrios Halkos Martin Haller Roger Hammons Redha Hamouche Joël Hancq Markus Hannula Lars Kai Hansen Alfred Hanssen Zaïd Harchaoui Md. Kamrul Hasan Anders Hast Thomas Haustein Rajesh M. Hegde Cornelius Hellge Richard C. Hendriks Conor Heneghan Ariane Herbulot Ewa Hermanowicz Jürgen Herre Ryan Hersey Otto Heunecke Richard Heusdens Ulrich Heute Heidi Hindberg Keigo Hirakawa Teemu Hirsimäki Václav Hlaváč Franz Hlawatsch Dominic K. C. Ho Aykut Hocanin William S. Hodgkiss Lars Höhmann Robert Höldrich Taras Holotyak Martin Holzer James R. Hopgood Adrian Hornsby Shahram Hosseini Ahmad Reza Hosseini-Yazdi Sébastien Houcke Yu-Tao Hsieh Hao Hu Gang Hua Heng Huang

Ke Huang Yufei Huang Abir Hussain Giovanni Iacovitti Jérôme Idier Masaaki Ikehara Alexander Ilin Cornel Ioana Ignasi Iriondo Prakash Ishwar Dan Istrate Alexei V. Ivanov Andrzej Izworski Jacek Izydorczyk Michael Jachan Philip JB Jackson Andreas Jakobsson Joakim Jaldén André Jalobeanu Jiří Jan Andrzej Janczak Magnus Jansson Jesper Jensen Ian Jermyn Michael Joham Hakan Johansson Don H. Johnson Philippe Joly Shiv Dutt Joshi Emil Jovanov Peter Jung Volker Jungnickel Markku J. Juntti Peter Kabal Tomasz Kacprzak Jari P. Kaipio Thomas Kaiser Markus Kallinger Hari Kalva Farzad Kamalabadi Marek Kamiński Karl Dirk Kammeyer Umasankar Kandaswamy Alexander Kaps Pantelis Karamolegkos Matti Karjalainen Stavros A. Karkanis Wolfgang Karner Damian Karwowski Rajgopal Kasi Andrzej Kasiński Włodzimierz Kasprzak Aggelos K. Katsaggelos Dimitris Katselis André Kaup Pasi Kauppinen Arata Kawamura Michał Kawulok Marián Képesi Andy W. H. Khong

Preben Kidmose Michel Kieffer Taewhan Kim Nick G. Kingsbury Keisuke Kinoshita Heiner Kirchhoffer Hagai Kirshner Josef Kittler Anssi Klapuri John Kleider Bastiaan Willem Kleijn Artur Klepaczko Jörg Kliewer Mark Kliger Sven Klomp Bastian Knerr Kevin H. Knuth Heinz Köppl Eleftherios Kofidis Ted Chi-Wah Kok Anil C. Kokaram Kostas Kokkinakis Géza Kolumbán Lisimachos Paul Kondi Ahmet Kondoz Adams Kong Janusz Konrad Ender Konuko˘glu Piotr Korbel Ulrike Korger Przemysław Korohoda Jukka Kortelainen Markus Koskela Miika Koskinen Bożena Kostek Jayesh Kotecha Constantine Kotropoulos Margarita Kotti Denis Kouamé Yevgeni Koucheryavy Oleksiy Koval Adam Kowalewski Michal Kozubek Jerzy Kołakowski Kidiyo Kpalma Emine Krichen Mohamed Krini Raghu Krishnamoorthi Silko-Matthias Kruse Denis Kubasov Heiner Kuhlmann Sławomir Kula Vibhor Kumar Anton Kummert Deepa Kundur Achim Kuntz Ragip Kurceren Ercan Kuruoğlu Azadeh Kushki Victor Man-Wai Kwan

Bogdan Kwolek Dimosthenis Kyriazis Fabrice Labeau Lionel Lacassagne Bernard Lacaze Jean-Louis Lacoume Míguel Ángel Lagunas Hernández Yeong-Kang Lai Klio Lakiotaki Danai Laksameethanasan Aris S. Lalos David S. Lalush Kin-Man Lam Andreas Lang Elmar Wolfgang Lang Rafał Lange Jan Larsen Erik Larsson Pascal Larzabal Samson Lasaulce Phooi Yee Lau Guillaume Lavoué Ngai-Fong Law Onoriu Lazar Cédric Le Barz Patrick Le Callet Gaëtan Le Guelvouit Nicolas Le Bihan Fritz Lebowsky Gottfried Lechner Hyuk-Jae Lee Jeong Woo Lee Miriam Leeser Frédéric Lefèbvre Dominique Leger Eric A. Lehmann Yee Hong Leung Geert Leus Rahim Abdul Leyman Olivier Lezoray Chunjian Li Heng Li Min Li Shujum Li Athanasios Liavas Aristidis C. Likas Livio Lima Chia-Wen Lin Han Lin Zhiping Penn Lin Joakim Lindblad Tamás Linder Adam T. Lindsay Cong Ling Antonios Litke Kejing Liu Judith Liu-Jimenez Joan Llach Kam W. Lo Heiner Löllmann

Elena-Simona Lohan Philip C. Loizou Marco J. Lombardi David G. Long José A. López-Salcedo Roberto López-Valcarce Philippe Loubaton Siow Yong Low Ting Lu Wu-Sheng Lu David García Luengo Peter Lüthi Marco Luise Rastislav Lukac Nils Löfgren Adam Łuczak Ewa Łukasik Thomas Mader Thomas Magesacher Enrico Magli Emanuele Maiorana Dirk Maiwald Ryszard Makowski Mladen Males Jean-Philippe Malkasse Daniele Mangano Athanassios Manikas Ali Mansour Gian Luca Marcialis Marco Marcon Evgeny Margolis Jean-Luc Mari Tomasz Markiewicz Renfors Markku Liam Marnane Ferran Marqués Maurizio Martina Guido Masera Andrea Masini Samuel Matej Andrzej Materka Driss Matrouf Pavel Matula Frédéric Maussang Sławomir Maćkowiak Alan McCree Tim McGraw Stephen McLaughlin Steven W. McLaughlin Christoph Mecklenbräuker Klaus Meerkötter Christian Mehlführer Sylvain Meignier Massimiliano Melani Bill Melvin Gloria Menegaz Xiaolin Meng Umberto Mengali Andreas Menychtas Philipp Merkle

Christian Merkwirth Russell M. Mersereau Alfred Mertins Olivier Meste Xavier Mestre Joaquín Míguez Zbigniew Mikrut Simone Milani Wided Miled Ben Milner Ramón Ricós Miralles Shigeki Miyabe Teruyuki Miyajima Marc Moeneclaey Mehdi Mohseni Rafael Molina Wojciech Molisz Gianluca Monaci Enric Monte-Moreno Marc Moonen Christine Mooshammer Robin D. Morris Vassiliki Moschou Gabriele Moser Stefanos Mpelekos Marta Mrak Karsten Müller Bernie Mulgrew Mario Munich Steen Michael Munk Adrian Munteanu Boris Murmann Peter Murphy Aloys Mvuma Jani Mäntyjärvi Montse Nájar Tomohiro Nakatani Antonio Napolitano Shrikanth S. Narayanan Girish Nathaniel Narra Matthias Narroschke Youssef Nasser Iulian Nastac Marek Natkaniec Monica Navarro Patrick Naylor Kianoush Nazarpour Patrick Ndjiki-Nya Nedko H. Nedev Arye Nehorai Markus Neinhüs Olivia Nemethova Alessandro Neri Khalid Niazi Didier Nicholson Maciej Jan Niedźwiecki Mariusz Nieniewski Ioannis G. Nikolakopoulos Christophoros Nikou Rickard Nilsson

Tsuyoki Nishikawa Shotaro Nishimura Takanobu Nishiura Mark Nixon Kai Noponen Jörgen Nordberg Sven Nordholm Bo Nordin Zbigniew Nosal José Luis Nuñez-Yañez Ingela Nyströms Elizabeth O'Sullivan Bogusław Obara Humberto Ochoa Halil Özer Marek R. Ogiela Maciej Ogorzałek Jane Ojanen Krzysztof Okarma Mikko Oksanen Ruşen Öktem Christian Olivier Hannu Olkkonen Erol Önen Nobutaka Ono Davide Onofrio Soontorn Oraintara Luis G. Ordónez Maciej Orkisz Javier Ortega Roberto R. Osorio Stanisław Osowski Hisham Othman Björn Ottersten Marina Ottonello Abdeldjalil Ouahabi Mourad Ouaret Viktor Öwall Neyir Ozcan Serdar Özen Mehmet Tankut Özgen Tuomas Paatero Jyri Tapani Pakarinen Kalle Palomäki Henryk Palus Sudhakar Pamarti Costas Panagiotakis Georgios TH. Papadopoulos George Papakostas Zdzisław Papir Matteo Pardini Marek Parfieniuk Antonio Pascual Marek Pawełczyk William A. Pearlman Maciej Pedzisz Paul H. Peeling Soo-Chang Pei François Pellegrino Denis Pellerin

Johannes Peltola Henri Penttinen Fernando Pereira Rochelle Pereira Tanguy Pérennou Luis Pérez-Freire Ana I. Pérez-Neira Haim Permuter Jean-Christophe Pesquet Stefan W. Petrausch Dijana Petrovska-Delacrétaz Alexander Petrovsky Tomas Pevny Françoise Peyrin Hartmut R. Pfitzinger Dinh-Tuan Pham Marian Piekarski Gemma Piella Lorenzo Pieralisi Piotr Pietrzak Aggelos Pikrakis Ruven Pillay Li Ping Julien Pinquier Susann Pirttikangas Alessandro Piva Peter Planinšič William L. Plishker Barnabás Póczos Moshe Porat Javier Portilla Alexander Potchinkov Bozidar Potocnik K. M. M. Prabhu Ian Proudler Artur Przelaskowski Henning Puder William Puech Ville Pulkki Janne Pylkkönen Cai Qipeng Tai-fan Quan André Quinquis Ludovic Quintard M. G. Rabbat Rudolf Rabenstein Stanisław Andrzej Raczyński Jan Raethjen Ewaryst Rafajlowicz Cappelli Raffaele Mario Raffin Bhiksha Raj Umesh Rajashekar Alain Rakotomamonjy Yothin Rakvongthai Giovanni Ramponi Pradeep Ramuhalli Tharm Ratnarajah Andreas Rauber Chandra Kant Raut

Mika Rautiainen Wolfgang Rave Marc Realp Hari C. Reddy André Redert Carlo Regazzoni Ulrich Reiter Alexander Reiterer Branimir Reljin Reede Ren Steve Renals Alexandre Renaux Markku Renfors Jaume Riba Cédric Richard Gaël Richard Jonas Richiardi Vincent Ricordel Michal Ries Roberto Rinaldo Jukka Rinne Mirko Ristivojevic Alessandro Rizzi William J. J. Roberts Marie Roch Florian Römer Hermann Rohling Mirosław Rojewski Adrian Romiński Patrice Rondao Yue Rong Athanasios Rontogiannis Roman Rosipal Arun Ross Francisco Rubio Mikael Rudberg Javier Ruiz Hidalgo Markus Rupp Jan Řstergaard Harri Saarnisaari Claudio Sacchi Brian M. Sadler Josep Sala Alvarez Addisson Salazar Philippe Salembier Raul Sanchez-Reillo Eduardo Sanchez-Soto Frida Sandberg Saeid Sanei Pekka Sangi Steve Sangwine Bülent Sankur Andrés Santos Lleó Sergio Saponara Tapio Saramäki Augusto Sarti Janusz Sawicki Ali H. Sayed Gaetano Scarano Jeffrey D. Scargle

Gerald Schaefer Sam Schauland Peter Schelkens David Schmidt Gerhard Schmidt Günter Schmitt Richard R. Schultz Thorsten Schulz Heiko Schwarz Andreas Schweizer Hans-Dieter Schütte Michael Scordilis Iain Scott Gesualdo Scutari Dominik Sczcerba Mohammad Hosein Sedaaghi Ivan Selesnick Raj Senani Seda Şenay Hakan G. Şenel Lotfi Senhadji Luca Serafini Gholam-Ali Seraji Sebastiano Serpico Kalpana Seshadrinathan William A. Sethares Eric Setton Serdar Sezginer Giorgos Sfikas Vinod Sharma Andrew Shearer Chung-Ching Shen Greg Showman Wiesław Sieńko Alberto Signoroni Radosław Sikora Jorge Silva Pierre Siohan Pradip Sircar Frank Sjöberg Władysław Skarbek Isaac Skog Clint Slatton Krzysztof Ślot Andrzej Śluzek Besma Smida Roman Śmierzchalski Bogdan Smołka Hing-Cheung So Rocco Soleti Gerald Sommer Xiaomu Song Martin Spiertz Susanna Spinsante Thomas Sporer Sascha Spors Ann Spriet Ashok N. Srivastava Patrick Stadelmann Rudolf Staiger

Ljubisa Stankovic Radomir S. Stankovic Ryszard Stasiński Nikolce Stefanoski Peter Steffen Mikael Sternad Bob Stewart Peter Stoica Paweł Strumiłło Tilo Strutz Michał Strzelecki Yannis Stylianou Bartłomiej Sulikowski Zhenan Sun Wonyong Sung Myung Hoon Sunwoo Luca Superiori Lennart Svensson Stina Svensson A.Lee Swindlehurst Przemysław Sypka Marek Szczepański Leif Sörnmo Paweł Śniatała Joseph Tabrikian Ryszard Tadeusiewicz Marco Tagliasacchi Yasushi Takatori Mario Tanda Oguz Tanrikulu Jarno M. A. Tanskanen Andrzej Tarczynski Georg Tauböck Murat A. Tekalp Livio Tenze Philippe Thévenaz Jean-Philippe Thiran Nadège Thirion-Moreau Sergios Thodoridis John S. Thompson Christophe Tillier Tammam Tillo Luca Timmoneri Massimo Tistarelli Nicolas Tizon Roberto Togneri Tuukka Toivonen Andrea M. Tonello Jeffrey S. Tongue Mats Torkelson Bruno Torrésani Renata Tourinho Jean Yves Tourneret Wojciech Trąbka Linh Tran Hoai Carlos Manuel Travieso-González Alain Tremeau Mahdi Triki Željen Trpovski Andrea Trucco

Patricia Maria Trujillo Joel H. Trussell Chien-Cheng Tseng Stefano Tubaro Rodica A. Tuduce Zekeriya Tüfekçi Ertem Tuncel John Tuthill Wojciech Tylman George Tzanetakis Panagiotis Tzionas Osman Nuri Uçan Francesca Uccheddu Dariusz Uciński Anna Ukovich Ali Hakan Ulusoy Olli Vainio Vesa Välimäki Mikko Valkama Dimitri Van De Ville Koen Van Den Abeele Annemie Van Hirtum Pierre Vandergheynst Allert VanZelst Peter Vary Libor Váša Alex O. Vasilescu Athanasios Vasilopoulos Raymond Veldhuis Jörg Velten Mahesh Vemula Fabio Verdicchio Alessandro Verri Dimitrios Ververidis Luminita Aura Vese Mark Vesterbacka Mats Viberg Ari Viholainen José Emilio Vila Forcén Pedro Gómez Vilda Renato F. Villán S. Pramod Viswanath Giorgio M. Vitetta Francesco Maria Vitullo Liviu Vladutu An Vo Ba-Tuong Vo Christian Vogel Sviatoslav Voloshynovskiy Sergiy A. Vorobyov Patrik Wahlberg Dirk Waldhauser Jacqueline Walker John MacLaren Walsh Junfeng Wang Wenwu Wang Zhisong Wang Zhou Wang Lars Wanhammar Shinji Watanabe

Robert Weber Joachim Wehinger Qingqing Wei Oomke Weikert Stephan Weiss Jens Wellhausen Markus Wenk Karl Werner Krzysztof Wesołowski Hannes Wettig Paul F. Whelan Cecilia Whitaker Nick P. Whiteley Risto Wichman Bogdan Więcek Mathias Wien Pascal Wiggers Frans M. J. Willems Simon Wilson Andreas Wilzeck Marcus Windisch Martin Winken Ingo Wolf Patrick Wolfe Dennis M.-L. Wong Kainam Thomas Wong Damon L. Woodard Wei Lee Woon Qiu Wu Renbiao Wu Carolina E. Å. Wählby Jinjun Xiao Yegui Xiao Zixiang Xiong Dong Xu Zhemin Xu Zhengyuan (Daniel) Xu Liuqing Yang Ming-Hsuan Yang Wenbing Yao Philippe G. Young Masahiro Yukawa Stefanos Zafeiriou Yuriy Zakharov Giuseppe Zamuner Santiago Zazo Bello Vesna Zeljkovic Thomas Zemen Azzedine Zerguine Andre Zeug Engin Zeydan Chengjin Zhang Li Zhang Li X Zhang Qianni Zhang Yonggang Zhang Meng Zhao Qijun Zhao Tong Zhao Daidi Zhong

Shengli Zhou Xu ( Judy ) Zhu Andreas Ziehe Tomasz Zieliński Ali Zifan Mariusz Ziółko Zoran Zivkovic Dmitry Znamenskiy Nizar Zorba Michele Zorzi Abdelhak M. Zoubir Steeve Zozor Mariusz Zubert Manuel Rosa Zurera

Plenary Lectures

MULTI-WAY BLIND SOURCE SEPARATION USING NONNEGATIVE MATRIX FACTORIZATION AND SPARSE COMPONENT ANALYSIS

Andrzej Cichocki RIKEN Laboratory, Japan

Nonnegative matrix factorization (NMF), Non-negative tensor factorization (NTF), parallel factor nalysis PARAFAC and TUCKER models with non-negativity constraints have been recently proposed as promising sparse and quite efficient representations of signals, images, or general data. From a viewpoint of multidimensional data analysis, NTF is very attractive because it takes into account spatial and temporal correlations between variables more accurately than 2D matrix factorizations, such as NMF or ICA and it provides usually sparse common factors or hidden (latent) components with physical or physiological meaning and interpretation. In this talk we review several general and flexible models with multilayer or recurrent structures. Especially, we discuss 3D tensors (also known as n-way arrays or multidimensional arrays) factorizations and discuss some potential applications ranging from neuroscience and bioinformatics. Application to biomedical signal processing and Brain Machine Interface (BMI) will be also briefly presented. We will present a new 3D tensor modeling (decomposition/factorization) approach and associated learning algorithms in applications to multi-way Blind Source Separation (BSS), multidimensional data analysis, and sparse image representations. Using generalized cost functions (alpha and beta divergences), we will present derivation and practical implementations of three classes of algorithms: Multiplicative, Fixed Point Alternating Least Squares (FPALS) and Alternating Interior-Point Gradient (AIPG) algorithms. Some of the proposed algorithms are characterized by improved efficiency and convergence rates and can be applied with various distributions of data and additive noise. We discuss various cost functions used in information theory, which allows us to obtain generalized forms of learning algorithms. We have confirmed by extensive simulations that our multilayer NTF approach with multi-start initializations improves performance of the proposed algorithms if a specific model is approximately valid. In this talk, we will also discuss briefly some alternative approaches and algorithms for blind signal decomposition, especially for ICA, and SCA in order to estimate unknown sources signals, to perform feature extraction, dimension reduction and object recognition, remove artifacts and denoising of multi-modal, multi-sensory data.

Andrzej Cichocki received the M.Sc. (with honors), Ph.D. and Dr.Sc. (Habilitation) degrees, all in electrical engineering. from Warsaw University of Technology (Poland). Since 1972, he has been with the Institute of Theory of Electrical Engineering, Measurement and Information Systems, Faculty of Electrical Engineering at the Warsaw University of Technology, where he obtained a title of a full Professor in 1995. He spent several years at University Erlangen-Nuerenberg (Germany), at the Chair of Applied and Theoretical Electrical Engineering directed by Professor Rolf Unbehauen, as an Alexander-von-Humboldt Research Fellow and Guest Professor. In 1995-1997 he was a team leader of the laboratory for Artificial Brain Systems, at Frontier Research Program RIKEN (Japan), in the Brain Information Processing Group. He is currently the head of the laboratory for Advanced Brain Signal Processing, at RIKEN Brain Science Institute (JAPAN) in the Brain-Style Computing Group directed by Professor Shun-ichi Amari. He is co-author of more than 200 technical papers and three internationally recognized monographs (two of them translated to Chinese): Adaptive Blind Signal and Image Processing (Wiley, April 2003-revised edition), CMOS Switched-Capacitor and Continuous-Time Integrated Circuits and Systems (Springer-Verlag, 1989) and Neural Networks for Optimizations and Signal Processing (Teubner-Wiley, 1994). He is Editor in Chief of International Journal Computational Intelligence and Neuroscience and Associate Editor of IEEE Transactions on Neural Networks.

CHANNEL - AWARE DISTRIBUTED ESTIMATION AND DETECTION USING WIRELESS SENSOR NETWORK

Georgios B. Giannakis University of Minnesota, USA

Outline of Topics: 1. Motivation and Context 1a. Energy and Bandwidth Constraints 1b. Applications 2. Distributed Detection and Estimation 2a. Universal and Channel-Aware Detection 2b. Parameter Estimation and Tracking 2c. Dimensionality Reduction and Compression 2d. Performance and Distortion-Rate Analyses 3. Wireless Communication Issues 3a. Synchronization Algorithms 3b. Channel-Aware Detection and Estimation 3c. Multiple Access and Resource Allocation 4. Networking Issues 5. Summary and Future Directions

G. B. Giannakis (Fellow'97) received his Diploma in Electrical Engr. from the Ntl. Tech. Univ. of Athens, Greece, 1981. From 1982 to 1986 he was with the Univ. of Southern California (USC), where he received his MSc. in Electrical Engineering, 1983, MSc. in Mathematics, 1986, and Ph.D. in Electrical Engr., 1986. Since 1999 he has been a professor with the ECE Department at the Univ. of Minnesota, where he now holds an ADC Chair in Wireless Telecommunications. His general interests span the areas of communications, networking and statistical signal processing - subjects on which he has published more than 250 journal papers, 450 conference papers, two edited books and two research monographs. Current research focuses on diversity techniques, complex-field and space-time coding, multicarrier, cooperative wireless communications, cognitive radios, cross-layer designs, mobile ad hoc networks, and wireless sensor networks.

G. B. Giannakis is the (co-) recipient of six paper awards from the IEEE Signal Processing (SP) and Communications Societies including the G. Marconi Prize Paper Award in Wireless Communications. He also received Technical Achievement Awards from the SP Society (2000), from EURASIP (2005), a Young Faculty Teaching Award and the G. W. Taylor Award for Distinguished Research from the University of Minnesota. An IEEE Fellow he has served the IEEE in various posts.

REINVENTING COMPRESSION: THE NEW PARADIGM OF DISTRIBUTED VIDEO CODING

Bernd Girod Stanford University, USA

Distributed coding is a new paradigm for video compression, based on Slepian and Wolf's and Wyner and Ziv's information-theoretic results from the 1970s. This talk reviews the recent development of practical distributed video coding schemes. Wyner-Ziv coding, i.e., lossy compression with receiver side information, enables low-complexity video encoding where the bulk of the computation is shifted to the decoder. Since the interframe dependence of the video sequence is exploited only at the decoder, an intraframe encoder can be combined with an interframe decoder. Wyner-Ziv coding is also naturally robust against transmission errors and can be used for joint source-channel coding by protecting the signal waveform rather than a compressed bit-stream. It thus achieves graceful degradation under deteriorating channel conditions without a layered signal representation. Besides low-complexity encoding and robust transmission, the distributed coding paradigm enables novel solutions to diverse problems ranging from coding for random access to media authentication to compression of encrypted signals.

Bernd Girod is Professor of Electrical Engineering and (by courtesy) Computer Science in the Information Systems Laboratory of Stanford University, California. He was Chaired Professor of Telecommunications in the Electrical Engineering Department of the University of Erlangen-Nuremberg until 1999. His research interests are in the areas of video compression and networked media systems, and he has published over 400 conference and journal papers, as well as 5 books. Professor Girod has been involved with several startup ventures as founder, director, investor, or advisor, among them Polycom (Nasdaq:PLCM), Vivo Software, 8x8 (Nasdaq: EGHT), and RealNetworks (Nasdaq: RNWK). Since 2004, he serves as the Chairman of the new Deutsche Telekom Laboratories in Berlin. He received the Engineering Doctorate from University of Hannover, Germany, and an M.S. Degree from Georgia Institute of Technology. Prof. Girod is a Fellow of the IEEE and a member of the German Academy of Sciences (Leopoldina). He received the 2002 EURASIP Best Paper Award, the 2004 EURASIP Technical Achievement Award, and the 2007 IEEE Multimedia Communication Best Paper Award.

EQUALIZATION OF MIMO CHANNELS

John G. Proakis and Patrick Amihood

Department of Electrical and Computer EngineeringUniversity of California at San Diego

La Jolla, CA 92093, USAemail: jproakis@ucsd.edu

ABSTRACTWe consider the equalization of multiple-input, multiple-

output (MIMO) wireless communication systems that employmultiple transmit and receive antennas to increase the datarate and achieve signal diversity in fading multipath chan-nels. Two scenarios in the equalization of MIMO systemsare treated. The first is a point-to-point MIMO system inwhich the channel characteristics are known at the receiveronly and, hence, the equalization is performed at the receiver.The second is a point-to-multipoint (broadcast) MIMO sys-tem in which the channel characteristics are known at thetransmitter. In this case, the equalization is performed atthe transmitter. Both linear and nonlinear equalization algo-rithms are treated.

1. INTRODUCTION

In this paper, we consider multiple-input, multiple-output(MIMO) wireless communication systems that employ mul-tiple transmit and receive antennas to increase the data rateand achieve signal diversity in fading multipath channels.The performance of MIMO systems is degraded by two typesof interference. One is intersymbol interference (ISI) dueto channel multipath dispersion. The other is cross-talk orinterchannel interference due to the simultaneous transmis-sions from the multiple transmit antennas. The focus of thepaper is on equalization and detection algorithms for mitigat-ing these two types of interference.

Two scenarios in equalization of MIMO systems aretreated, as shown in Figure 1. The first is a point-to-pointMIMO system in which the channel characteristics are un-known at the transmitter. In this case, the equalization and in-terference mitigation is performed at the receiver, which es-timates the channel characteristics from pilot signals that aretransmitted by the multiple transmit antennas. The secondscenario is a point-to-multipoint (broadcast) MIMO trans-mission system in which the channel characteristics are as-sumed to be known at the transmitter. In this case, the equal-ization and interference mitigation is performed at the trans-mitter. Both linear and nonlinear interference mitigation al-gorithms are considered.

2. POINT-TO-POINT MIMO SYSTEM

The general configuration for a point-to-point MIMO com-munication system is shown in Figure 2. For simplicity, weassume that M-ary PSK or QAM modulation is used. Thedata is encoded and interleaved by a pseudo-random inter-leaver of some appropriately chosen length. A block of NTcoded data symbols is converted from serial to parallel andfed to NT identical modulators, where each modulator is con-

nected to a separate antenna. The transmitted signals arereceived by NR receive antennas whose signals are fed toseparate demodulators. The outputs of the demodulators arepassed to a detector, converted from parallel to serial, de-interleaved and decoded. In the following, we refer to such asystem as an (NT , NR) MIMO system.

We assume that the channels between each transmitand receive antenna are statistically independent, slowlytime-varying Rayleigh fading multipath channels which areknown to the NR receivers. We also assume that the differ-ence in propagation times of the signals from the NT transmitantennas to the NR receive antennas are small relative to thesymbol duration T , so that for practical purposes, the signalsfrom the NT transmit antennas to any receiving antenna aresynchronous. The channel between each transmit and receiveantenna, including transmitter and receiver filters, is modeledas a linear tapped-delay-line filter having finite duration im-pulse response that spans L successive symbols. Thus, L−1is the span of the multipath spread and the span of the ISI insymbols.

First, let us consider an uncoded system. The opti-mal receiver for an (NT , NR) MIMO multipath channel is amaximum-likelihood sequence detector (MLSD) that jointlydetects the sequence of NT -dimensional data vectors that aretransmitted in successive symbol intervals. This receiverstructure is a generalization of the well-established single-input, single-output (SISO) MLSD, which is efficiently im-plemented by use of the Viterbi algorithm [1, 2].

The performance of the MLSD receiver in Rayleigh fad-ing with statistically independent additive white Gaussiannoise in the received signals of the NR antennas has beenevaluated in [3]. It is demonstrated in this paper that theMLSD receiver achieves a signal diversity of order LNR.That is, for a spatial multiplexing of order NT , the MLSDreceiver exploits the full diversity that is available from thechannel multipath components and the NR receive antennas.

When the number of transmit and receive antennas islarge, the computational complexity of the MLSD receivermay prohibit practical implementations. In such cases, ei-ther linear or decision-feedback equalizers , as illustrated inFigures 3 and 4, may be employed in place of the maximum-likelihood detector. Such equalizers generally yield subopti-mum performance compared to MLSD and do not exploit thefull signal diversity that results from channel multipath andmultiple receive antennas. The performance of linear anddecision-feedback equalizers for an (NT , NR) MIMO systemhas also been investigated in [3].

Let us now consider the use of coding in the (NT , NR)MIMO system shown in Figure 2, where the code is as-sumed to be a binary convolutional code and the interleaver

is a bit interleaver. Since the channel is modeled as a lin-ear, tapped-delay line filter of finite duration, the concatena-tion of the convolutional encoder, the interleaver, and eachchannel between a transmit antenna and a receive antennamay be viewed as a serially concatenated convolutional code(SCCC). To reduce the complexity at the receiver, we maydecode the constituent codes separately and iteratively, i.e.,we may perform turbo equalization [4]. This approach yieldsnear optimal performance. A receiver structure that per-forms iterative equalization and decoding is illustrated inFigure 5. As shown, the equalizer coefficients may be op-timized in each iteration based on the maximum a posteriori(MAP) probability criterion and the decoder is also a MAPdecoder. The computations performed by the MAP equalizerand MAP decoder are described in some detail in [3].

3. POINT-TO-POINT MIMO SYSTEMSEMPLOYING SPREAD SPECTRUM SIGNALS

MIMO systems have been proposed for use in cellularCDMA data communication systems with the goal of provid-ing high data rates through spatial multiplexing and increasedspatial diversity obtained via multiple receive antennas. Onesuch application is proposed in the 3rd Generation Partner-ship Project (3GPP) called High-Speed Downlink Packet Ac-cess (HSDPA), which employs orthogonal spreading codesto multiplex multiple data sequences. A block diagram ofthe transmitter and receiver of such a MIMO system is shownin Figure 6. We note that by using Nc orthogonal spreadingsequences at each transmit antenna, we are multiplexing Ncsymbols that are transmitted simultaneously from each an-tenna. The Nc information-bearing spread spectrum signalsat each transmit antenna are further scrambled by multipli-cation with a pseudo-random sequence denoted as pi. Thesespread spectrum MIMO systems are called multicode sys-tems.

Time dispersion due to multipath causes interchip inter-ference which may be suppressed by use of chip-based equal-izers. In Figure 5, linear equalizers are used. In such a sys-tem, one of the orthogonal codes at each transmit antennais left unmodulated and is used as a pilot-signal at the re-ceiver of each receive antenna to adjust the coefficients ofthe equalizers. The mean-squared error (MSE) criterion is asimple and implementable performance metric for optimiz-ing the coefficients of the equalizers. Iterative algorithmssuch as the LMS and RLS algorithms have been investigatedfor adaptively adjusting these coefficients [5-8].

In general a DFE offers the advantage of improved per-formance compared to a linear equalizer. In a spread spec-trum system, the feedforward filters should process the re-ceived signals at the chip rate. However, the feedback filtersshould not be operated at the chip rate, because chip-level de-cisions are very unreliable. Figure 7 illustrates a DFE struc-ture in which the feedback filter operates at the symbol rateand the feedforward filter is operated at some multiple of thechip rate.

Another DFE structure for an HSDPA MIMO system,called a linear-aided DFE is described in [9]. This linear-aided DFE employs a two-pass operation through the dataas shown in Figure 8 where linear equalization is used asa first stage and DFE is used as a second stage. The lin-ear equalizers operate at the chip rate and perform chip-levelequalization, using pilot signal chip sequences for adaptively

adjusting the coefficients of the equalizers. The outputs ofthe equalizers are appropriately combined, despread, and thesymbols are detected. The detector output constitutes tenta-tive decisions which are used to regenerate the spread spec-trum signals that are fed to the feedback filter of the DFE.In this configuration, the feedback filter is operated at thechip rate. The performance of this linear-aided DFE has beenevaluated for an (NT , NR) = (4, 4) MIMO system in [9] andcompared with that of linear equalizers.

4. POINT-TO-MULTIPOINT MIMO SYSTEM

In the previous sections, we considered point-to-point MIMOsystems in which the equalization of intersymbol interfer-ence and interchannel interference was performed at the re-ceiver. In this section, we consider a point-to-multipoint(broadcast) MIMO system which transmits data simultane-ously to multiple users that are geographically distributed.The transmitter is assumed to employ NT antennas to trans-mit data to K receivers, where NT ≥K. Each user is assumedto have a receiver with one or more receiving antennas. Thisscenario applies, for example, to the downlink of a wirelesslocal area network. The distinguishing feature of the MIMObroadcast system is that the receivers, which are geographi-cally separated, do not employ any coordination in process-ing the received signals.

In a MIMO broadcast system, there are two possible ap-proaches for dealing with the multiple-access interference(MAI) resulting from the simultaneous transmission to mul-tiple users. One approach is to have each receiver employinterference mitigation in the recovery of its desired signal.In most cases, this approach is impractical because the userslack the processing capability and are constrained by the lim-ited energy resources inherent in the use of battery power.The alternative approach is to employ interference mitiga-tion at the transmitter, which possesses significantly moreprocessing capability and energy resources.

To mitigate the MAI at the transmitter, the transmittermust know the channel characteristics, typically, the chan-nel impulse response. This channel state information (CSI)may be obtained from channel measurements performed ateach of the receivers by means of received pilot signals sentby the transmitter. Then, the CSI must be sent to the trans-mitter. In such a scenario, the channel time variations mustbe relatively slow so that a reliable estimate of the channelcharacteristics is available at the transmitter. In some sys-tems, the uplink and downlink channels are identical, e.g.,the same frequency band is employed for both the uplink andthe downlink, but separate time slots are used for transmis-sion. This transmission mode is called time-division duplex(TDD). In TDD systems, the pilot signals for channel mea-surement may be sent by each of the users in the uplink. Inour treatment below, we assume that the CSI at the transmit-ter is perfect.

The suppression of MAI by means of transmitter process-ing is usually called signal precoding. Signal precoding at thetransmitter may take one of several forms, depending on thecriterion or the method used to perform the precoding. Thesimplest precoding methods are linear and are based on ei-ther the zero-forcing (ZF) criterion or the MSE criterion. Onthe other hand, there are nonlinear signal precoding methodsthat result in better system performance. First, we treat lin-ear precoding and, then, we describe a nonlinear precoding

method and compare their performance.

4.1 Linear Precoding of the Transmitted SignalsIn this section, for mathematical convenience, we assumethat each user has a single antenna and the number of re-ceivers (users) is K ≤ NT . It is also convenient to assumethat the channel is nondispersive. The communication sys-tem configuration is shown in Figure 9, where the precodingmatrix is denoted as AT . Hence, the received signal vector is

y = HAT s+η

where H is an K x NT matrix, AT is an NT x K precodingmatrix, s is an K x 1 vector and η is an K x 1 Gaussian noisevector. The matrix that eliminates the MAI at each receiveris generally given by the Moore-Penrose pseudo-inverse

H+ = HH(HHH)−1

Hence, the precoding matrix is

AT = αH+ (1)

where α is a scale factor that is selected to satisfy the totaltransmitted power allocation, i.e., ‖ AT s ‖2= P. Thus, theprecoding matrix in (1) allows the individual users to recovertheir desired symbols without any interference from the sig-nals transmitted to the other users. We also observe that inthe special case where K = NT , AT = αH−1, so that theprecoding matrix is proportional to the inverse channel ma-trix. This constitutes a zero-forcing equalizer implementedat the transmitter.

Figure 10 illustrates the error rate performance of thezero-forcing precoder obtained via Monte Carlo simulationfor K = NT = 4, 6 and 10 and QPSK modulation. The chan-nel matrix elements are complex-valued i.i.d. zero-meanGaussian random variables with unit variance. We observethat the error rate increases with an increase in the number ofusers. We attribute this deterioration in performance to theill-conditioning of the channel matrix H. This is the majordrawback with the zero-forcing precoder.

If we relax the condition that the interference be zeroat all the receivers, the performance degradation can be re-duced. This can be accomplished by using the linear MSEcriterion in the design of the precoding matrix AT . Thus, weselect AT to minimize the cost function

J(AT ,α) = arg minα,AT

E∥∥∥∥

1α

(HAT s+η)− s∥∥∥∥

2

(2)

subject to the transmitted power allocation ‖AT s‖2 = P andwhere the expectation in (2) is taken over the noise and signalstatistics. The solution to the MSE criterion is the precodingmatrix

AT = αHH(HHH +βI)−1

where α is the scale factor that is selected to satisfy the powerallocation and β is defined as a loading factor, which whenselected as β = K/P maximizes the signal-to-interference-plus-noise ratio (SINR) at the receiver [10].

The error rate performance of the MMSE linear precoderobtained by Monte Carlo simulation in a frequency nonse-lective Rayleigh fading channel is illustrated in Figure 11 for

K = NT = 4,6 and 10. We observe that the error rate perfor-mance improves slightly as the number of users K increasesand that it exceeds the performance of the zero-forcing pre-coder.

4.2 Nonlinear Precoding of the Transmitted Signals:The QR DecompositionWhen the transmitter knows the interference caused on otherusers by the transmission of a signal to any particular user,the transmitter can design signals for each of the other usersto cancel the interference. The major problem with such anapproach is to perform the interference cancellation withoutincreasing the transmitter power. This same problem is en-countered in decision-feedback channel equalization of a sin-gle user, where the feedback filter of the DFE is implementedat the transmitter. In that case, when the range of the dif-ference between the desired transmitted symbol and the ISIexceeds the range of the desired transmitted symbol, the dif-ference is reduced by subtracting an integer multiple of 2Mfor M-ary PAM, where [−M,M) is the range of the desiredtransmitted symbol. This same nonlinear precoding methodcalled Tomlinson-Harashima precoding [11, 12], can be ap-plied to the cancellation of the MAI in a MIMO broadcastcommunication system [13-15].

Figure 12 illustrates the precoding operations for theMIMO multiuser system. The channel impulse response be-tween the ith transmit antenna and the receive antenna of thekth user is given by

hki(t) =L−1

∑l=0

h(l)ki δ (t− lT )

where L is the number of multipath components in thechannel response, T is the symbol duration, and h(l)

ki is thecomplex-valued channel coefficient for the lth path. Thechannel coefficients {h(l)

ki } are known at the transmitter andare realizations of i.i.d. zero-mean, circularly symmetriccomplex Gaussian random variables with variance

E[|h(l)

ki |2]

=1L

, ∀k, i and l

It is convenient to arrange these channel coefficients for thelth path in a K x NT matrix H(l), where [H(l)]ki = h(l)

ki , i =1,2, ...,NT , k = 1,2, ...,K.

The MAI cancellation is facilitated by use of the QR de-composition of the channel matrix H(0). Thus, we express[H(0)]H as

[H(0)]H = QR

where Q is an NT x K matrix, such that QQH = I, and Ris a K x K upper triangular matrix with diagonal elements{rii}. Based on this decomposition of [H(0)]H , the signal tobe transmitted is precoded with the matrix transformation

W = QA

where A is a K x K diagonal matrix with diagonal elements1/rii , i = 1,2, ...,K. The {rii} are real and positive [16]. Thematrix P = pI is a diagonal K x K matrix that is used simply

for scaling the power of the transmitted signal and resultsin equal SNR for all users. Therefore, we have an effectivechannel matrix of the form

H(0)WP = [QR]HQAP= pRHA

We note that RHA is a K x K lower triangular matrix withunit diagonal elements. As a result, user k sees multiple ac-cess interference from users 1,2, ...,k−1. We also note thatthe effective channel matrix H(0)W = RHA will have fullrank K, provided that NT ≥ K.

By reducing this channel matrix to a lower triangular ma-trix, we can now subtract the interference at the transmitterthat each user would normally observe at their respective re-ceiver. Thus, when the channel adds the same interferenceto the transmitted signal, the received signal at each receiverwould be free of interference. Taking advantage of the lowertriangular matrix structure, successive interference cancella-tion is performed with the feedback filter defined by the ma-trix

B = [I−H(0)W,−H(1)W , −H(2)W, ...,−H(L−1)W]

where the matrix (I−H(0)W) is used to cancel the interfer-ence due to the other users that arises in the current symbolinterval, and the terms −H(1)W−H(2)W, ...,−H(k−1)Ware used to cancel the interference due to previous symbols.

To ensure that the subtraction of the interference termsdo not result in an increase of transmitter power, we use themodulo operator, as in Tomlinson-Harashima precoding, tolimit the range of the signal to the boundaries of the signalconstellation. Thus, the output of the modulo operators forthe nth symbol vector, as shown in Figure 12 is (for squareQAM constellations)

x(n) = mod2√

M[s(n)+Bx(n)]= s(n)+Bx(n)−2

√Mzx(n)

where the modulo operation is performed on each real andimaginary component of the vector [s(n) +Bx(n)],x(n) isthe K x 1 vector at the output of the modulo operator, s′(n)is the K x 1 data vector, x(n) is defined as

x(n) = [x(n)T , x(n−1)T , x(n−2)T , ..., x(n− (L−1))T ]T

and zx(n) is an K x 1 vector with complex-valued compo-nents that take on integer values, determined by the con-straint that the real and imaginary components of x(n) fallin the range of [−√M,

√M). Therefore, the transmitted sig-

nal vector is expressed as

s′(n) = WPx(n)= pWx(n)

and the received signal vector is

r(n) = pL−1

∑i=0

H(i)Wx(n− i)+η(n)

and, hence,

P−1r(n) = x(n)+(H(0)W− I)x(n)

+L−1

∑i=1

H(i)Wx(n− i)+η ′(n) (3)

By substituting for B and x(n) in (3), it follows that

P−1r(n) = s(n)+η ′(n)−2√

Mzx(n)

Consequently, the MAI and ISI are cancelled perfectly, re-sulting in the test statistics for the nth symbol vector as

y(n) = mod2√

M

[1pr(n)

]

4.2.1 Optimum Ordering of the Decentralized Receivers

The ordering of the K decentralized receivers affects the con-struction of the K x NT channel matrix H(0). There are K!possible column permutations of [H(0)]H and, hence, there isone QR decomposition associated with each permutation. Inturn, there are K! transformation matrices W = QA, eachof which requires a different transmit power. In order tominimize the total transmit power, it is necessary to searchover all the column permutations. Such an exhaustive searchprocedure is computationally time-consuming, except for asmall number of users. A method for simplifying the searchfor the optimum ordering is described in [17].

The error rate performance of the QR decompositionmethod described above has been evaluated in [18, 19]. Fig-ure 13 illustrates the symbol error probability as a function ofthe SNR (total transmitted signal power over all antennas di-vided by N0) for QPSK modulation, L = 1, 2 and NT = K = 2.The Monte Carlo simulation results are also illustrated. Thesimulation results are obtained by transmitting 1,000 datasymbols over each of 10,000 channel realizations.

Figure 14 shows analytical results of the symbol errorrate performance for QPSK with L = 1 (flat fading), K = 2and NT = 2,3,4. We observe that the system performanceimproves with an increase in the number of transmit anten-nas, which reflects the benefit of spatial diversity.

Figure 15 shows simulation results comparing the errorrate performance of the linear zero-forcing (ZF) and min-imum MSE (MMSE) precoding methods with the QR de-composition method for QPSK modulation with L = 1, andK = NT = 4. We observe that the performance of the QR de-composition method is better than that of the linear precodersat high SNR, but poorer at low SNR. It should be noted thatthe improvement in performance of the QR decompositionmethod at high SNR is obtained at a significantly higher com-putational complexity compared with the linear precoders.

5. CONCLUDING REMARKS

In this paper we considered equalization and interferencesuppression in point-to-point and point-to-multipoint (broad-cast) MIMO wireless communication systems. Channelequalization at the receiver is especially suitable in point-to-point MIMO channels and effective equalization algorithmshave been devised which are essentially a generalization ofequalization algorithms for single-input, single-output sys-tems.

In point-to-multipoint MIMO systems, we focused on anonlinear equalization method performed at the transmitterthat employs the QR decomposition and illustrated its errorrate performance. In addition to this nonlinear method forMIMO channel equalization at the transmitter, other methodsbased on vector precoding and lattice reduction have beenconsidered in the literature [20-25]. These methods will bedescribed in the oral presentation.

6. ACKNOWLEDGEMENT

This work was supported by the US Army Research Officeunder MURI grant number W911NF-04-1-0224.

REFERENCES[1] J. G. Proakis. Digital Communications. New York:

McGraw-Hill, 4th Ed., 2001.[2] G. D. Forney, “Maximum-likelihood Sequence Estima-

tion of Digital Sequences in the Presence of IntersymbolInterference,” IEEE Trans. Inf. Theory, vol. 18, pp. 363-378, May 1972.

[3] B. A. Bjerke and J. G. Proakis, “Equalization and Decod-ing for Multiple-Input Multiple-Output Wireless Chan-nels,” EURASIP J. on Applied Signal Proc., vol. 3, pp.2249-266, March 2002.

[4] C. Douillard, M. Jezequel, C. Berrou, A. Picart, P. Di-dier, and A. Glavieux, “Iterative Correction of Intersym-bol Interference: Turbo-equalization,” European Trans.on Telecommun., vol. 6, no. 5, pp. 507-511, Sept-Oct1995.

[5] L. Mailaender, “Linear MIMO Equalization for CDMADownlink Signals with Code Reuse,” IEEE Trans. Wire-less Commun., vol. 4, no. 5, pp. 2423-2434, Sept 2005.

[6] C. Mehlfuhrer and M. Rupp, “A Robust MMSE Equal-izer for MIMO Enhanced HSDPA,” in Proc. 40th Asilo-mar Conference on Signals, Systems and Computers, Pa-cific Grove, CA, 2006.

[7] D. R. Morgan, “An Adaptive RLS MIMO Equalizer Al-gorithm for HSDPA,” in Proc. 40th Asilomar Conferenceon Signals, Systems and Computers, Pacific Grove, CA,2006.

[8] D. C. Garrett, L. M. Davis, and G. K. Woodward, “19.2Mbps 4 x 4 BLAST/MIMO Detector with Soft ML Out-puts,” IEE Electronics Letters, vol. 39, pp. 233-235, Jan-uary 2003.

[9] L. Mailaender and J. G. Proakis, “Linear-aided decision-. feedback equalization for the CDMA downlink,” inProc. 37th Asilomar Conference on Signals, Systems andComputers, Pacific Grove, CA, 2003.

[10] C. B. Peel, B. M. Hochwald, and A. L. Swindlehurst,“A Vector-Perturbation Technique for Near-CapacityMultiantenna Multiuser Communication - Part I: Chan-nel Inversion and Regularization,” IEEE Trans. Com-mun., vol. 53, pp. 195-202, Jan. 2005.

[11] M. Tomlinson,“A New Automatic Equalizer Employ-ing Modulo Arithmetic,” IEE Electronics Letters, vol. 7,pp. 138-139, Mar. 1971.

[12] H. Harashima and H. Miyakawa, “Matched-Transmission Technique for Channels with IntersymbolInterference,” IEEE Trans. Commun., vol. COM-20, pp.774-780, August 1972.

[13] R. F. H. Fischer, Precoding and Signal Shaping for Dig-ital Transmission, New York: John Wiley & Sons, Inc.,2002.

[14] R. F. H. Fischer, C. Windpassinger, A. Lampe, and J.B. Huber, “Space-Time Transmission Using Tomlinson-Harashima Precoding,” Proc. 4th Intern. ITG Conferenceon Source and Channel Coding, pp. 139-147, Berlin, Jan2002.

[15] C. Windpassinger, R. F. H. Fischer, T. Vencel, and J.B. Huber, “Precoding in Multi-antenna and Multi-userCommunications,” IEEE Trans. Wireless Commun., vol.3, pp. 1305-1316, July 2004.

[16] A. M. Tulino and S. Verdu. Random Matrix Theory AndWireless Communications. Now Publishers Inc., June 28,2004.

[17] G. J. Foschini, G. D. Golden, R. A. Valenzuela, P.W. Wolniansky, “Simplified processing for high spec-tral efficiency wireless communication employing multi-element arrays,” IEEE J. Selected Areas Commun., vol.17, pp. 1841-1852, Nov. 1999.

[18] P. Amihood, E. Masry, L. B. Milstein, and J. G. Proakis,“Analysis of a MISO Pre-BLAST DFE Technique forDecentralized Receivers,” Proc. 40th Asilomar Confer-ence on Signals, Systems and Computers, Pacific Grove,CA, November 2006.

[19] P. Amihood, E. Masry, L. B. Milstein, and J. G. Proakis,“Performance Analysis of a Pre-BLAST DFE Techniquefor MISO Channels with Decentralized Receivers,” ac-cepted in IEEE Trans. Commun., January 2007.

[20] B. M. Hochwald, C. B. Peel, and A. L. Swindlehurst,“A Vector-Perturbation Technique for Near-CapacityMultiantenna Multiuser Communication - Part II: Per-turbation,” IEEE Trans. Commun., vol. 53, pp. 537-544,Mar. 2005.

[21] G. Ginis and J. Cioffi, “Vectored Transmission for Dig-ital Subscriber Line Systems,” IEEE J. Selected AreasCommun., vol. 20, pp. 1085-1104, June 2002.

[22] C. Windpassinger, R. F. H. Fischer, and J. B. Huber,“Lattice-reduction-aided Broadcast Precoding,” IEEETrans. Commun., vol. 52, pp. 2057-2060, Dec. 2004.

[23] H. Yao and G. W. Wornell, “Lattice-Reduction-AidedDetectors for MIMO Communication Systems,” in Proc.Global Commun. Conf. (GLOBECOM-2002), Taipai,Taiwan, Nov. 2002.

[24] U. Erez and S. ten Brink, “Approaching the Dirty PaperLimit in Canceling Known Interference,” in Proc. 41stAllerton Conf. on Commun., Control, and Computing,Allerton, Il., Oct. 1-3, 2003.

[25] U. Erez and S. ten Brink, “A Close-to-Capacity DirtyPaper Coding Scheme,” IEEE Trans. Inf. Theory, vol. 51,pp. 3417-3432, Oct. 2005.

TX RX... ...

(a) Point-to-Point MIMO System

TX

RX

RX

RX

... ...

(b) Point-to-Multipoint(Broadcast) MIMO System

Figure 1: Two types of MIMO systems

Detector...

.

.

.

Modulator

Modulator

.

.

.

to

Converter

Parallel

Serial

Serial

to

Parallel

Converter

Encoder Interleaver

Data

DataDemodulator

Demodulator

Decoder Deinterleaver

Figure 2: Model of Point-to-Point MIMO CommunicationSystem

s1(n)

s2(n)

Samplerr1(t)

FIREqualizer

a11(n)

Sampler

Sampler

FIREqualizer

FIREqualizer

FIREqualizer

FIREqualizer

FIREqualizer

r2(t)

r3(t)

a12(n)

a21(n)

a22(n)

a31(n)

a32(n)

h11(τ ; t)

h32(τ ; t)

h22(τ ; t)

h31(τ ; t)

h21(τ ; t)

h12(τ ; t)

s1(n)

s2(n)

Figure 3: Signal demodulation with linear equalizers for an(NT = 2, NR = 3) MIMO system

Samplerr1(t)

Sampler

FeedbackFilter

Detector

FeedbackFilter

b11(n)

b12(n)

FeedbackFilterb21(n)

Detector

FeedbackFilterb22(n)

FIREqualizer

a11(n)

s2(n)

s1(n)

FIREqualizer

FIREqualizer

FIREqualizer

r2(t)

a12(n)

a21(n)

a22(n)

s2(n)

s1(n)

h11(τ ; t)

h12(τ ; t)

h21(τ ; t)

h22(τ ; t)

Figure 4: Signal demodulation with decision-feedbackequalizers for an (NT = 2, NR = 2) MIMO system

� �� �� �� �� �

� � � � � � � ����

� �� � � �

� �� � � ���

� � � � � � ��

��� � �� �� �� ��� � ��� � � �

� � � � �� ��� � � �! � � � � " ��

#

##

#

$%$&$

$%$&$$%$&$ $%$&$

' ' '

�

Figure 5: Iterative MAP Equalizer/Decoder for an (NT , NR)MIMO System

d1,Nc(t)

d1,1(t) p1(t)

dNt,1(t)

dNt,Nc(t)

pNt(t)

h1,1(t)

hNt,Nr(t)

hNt,1(t)

h1,Nr(t)

yNr(n)

y1(n)

v1,Nr(n)

vNt,1(n)

z1(n)

zNt(n)

...

......

...

...

s1,1(t)

sNt,Nc(t)

s1,Nc(t)

sNt,1(t)

1√

Tch

∫ (n+1)Tch

nTch(·) dt

...1√

Tch

∫ (n+1)Tch

nTch(·) dt

FIR EQ

FIR EQ

FIR EQ

FIR EQ

vNt,Nr(n)

v1,1(n)

(a) System Model

Symbol Detector

zi(n)

1L

L+(κ−1)−1

∑n=κ−1

1L

L+(κ−1)−1

∑n=κ−1

Symbol Detector

Zi,Nc(L)

Zi,1(L)

si,Nc(n− (κ−1))

si,1(n− (κ−1))

...

(b) Despreading and Symbol Detection

Figure 6: Multicode MIMO System

Linear

Equalizer

a1

∑( ) Detector

Detector

Feedback

Filter

b1

∑( )

s1

s2

c2

c1

p

Linear

Equalizer

Feedback

Filter

a2

b2

h2(t)

h1(t)

s1

s2

Chip-rate

Processing

Symbol-rate

c2

p

c1

Processing

Figure 7: DFE for a Multicode System

Signals

DespreadDetector

Respread

Signals

Feedback

Filters

Signals

DespreadDetector

D1

Delay

Delay

DNR

.

.

....

.

.

....

.

.

.

.

.

.

.

.

....

.

.

....

.

.

.

.

.

.

.

.

.

+

+

-

-

Input

Output

Symbols

Transmitter

Linear

Equalizers

Linear

Feedforward

Equalizers

Figure 8: Block diagram of Linear-Aided DFE for HSDPA

RX1

RX2

RXK

xs

s1

s2

sK

Linear

PrecoderChannel

AT

H

η

Figure 9: Model of MIMO Broadcast System EmployingLinear Precoding

0 5 10 15 20 25 3010

−4

10−3

10−2

10−1

100

SNR (dB)

Pro

babi

lity

of s

ymbo

l err

or

NT=K=10, 6, 4

Figure 10: Performance of ZF linear precoding with NT =K = 4,6,10.

0 5 10 15 20 25 3010

−4

10−3

10−2

10−1

100

SNR (dB)

Pro

babi

lity

of s

ymbo

l err

or

NT=K=4, 6, 10

Figure 11: Performance of MMSE linear precoding withNT = K = 4,6,10.

m2√

M(·)

m2√

M(·)

m2√

M(·)

...

...

x2(n)

x1(n)

...

B

s1(n)

s2(n)

sK(n) xK(n)

P W

...

...

1/p

m2√

M(·)

1/p

m2√

M(·)

y1(n)s′1(n)

s′2(n)

s′NT(n)

r1(n)

yK(n)rK(n)

Figure 12: Tomlinson-Harashima Precoding Applied to aMIMO System

0 5 10 15 20 25 3010

−3

10−2

10−1

100

SNR (dB)

Pro

babi

lity

of s

ymbo

l err

or

SimulationAnalytical

L=2

L=1

Figure 13: Performance of optimal ordered QR decomposi-tion with NT = K = 2 and L = 1, 2.

0 5 10 15 20 25 3010

−3

10−2

10−1

100

SNR (dB)

Pro

babi

lity

of s

ymbo

l err

or

NT=2

NT=3

NT=4

Figure 14: Performance of optimal ordered QR decomposi-tion with K = 2, L = 1, and NT = 2, 3, and 4.

0 5 10 15 20 25 3010

−4

10−3

10−2

10−1

100

SNR (dB)

Pro

babi

lity

of s

ymbo

l err

or

QR decomposition − optimal orderingMMSEZF

Figure 15: Comparison of the QR decomposition and thelinear precoders with NT = K = 4.

John Proakis (S'58-M'62-F'84-LF'99) received the BSEE from the University of Cincinnati in 1959, the MSEE from MIT in 1961 and the Ph.D. from Harvard University in 1967. He is an Adjunct Professor at the University of California at San Diego and a Professor Emeritus at Northeastern University. He was a faculty member at Northeastern University from 1969 through 1998 and held the following academic positions: Associate Professor of Electrical Engineering, 1969-1976; Professor of Electrical Engineering, 1976-1998; Associate Dean of the College of Engineering and Director of the Graduate School of Engineering, 1982-1984; Interim Dean of the College of Engineering, 1992-1993; Chairman of the Department of Electrical and Computer Engineering, 1984-1997. Prior to joining Northeastern University, he worked at GTE Laboratories and the MIT Lincoln Laboratory.

His professional experience and interests are in the general areas of digital communications and digital signal processing. He is the author of the book Digital Communications (New York: McGraw-Hill, 2001, fourth edition), and co-author of the books, Introduction to Digital Signal Processing (Upper Saddle River: Prentice Hall, 2007, fourth edition); Digital Signal Processing Laboratory (Englewood Cliffs: Prentice Hall, 1991); Advanced Digital Signal Processing (New York: Macmillan, 1992); Algorithms for Statistical Signal Processing(Upper Saddle River: Prentice Hall, 2002);Discrete-Time Processing of Speech Signals (New York: Macmillan, 1992, IEEE Press, 2000); Communication Systems Engineering, (Upper Saddle River: Prentice Hall, 2002, second edition); Digital Signal Processing Using MATLAB V.4 (Boston: Brooks/Cole-Thomson Learning, 2007, second edition); Contemporary Communication Systems Using MATLAB (Boston: Brooks/Cole-Thomson Learning, 2004, second edition); Fundamentals of Communication Systems (Upper Saddle River: Prentice Hall , 2005).

RECENT ADVANCES IN VIDEO CODING AND TRANSMISSION

Thomas Wiegand

Fraunhofer Institute for Telecommunications, Germany

Higher compression gains, improved error robustness, increased adaptability and more functionality in light of the continuous evolution of networks and end devices are the most challenging factors of modern video coding and transmission. This talk aims to address those aspects by discussing the current state of H.264/AVC video coding standardization and its upcoming extensions together with the technical ecosystem and related research. For H.264/AVC, the state-of-the-art is reviewed and new optimization approaches for encoders as well as new techniques for enhanced video compression are discussed. One new extension of H.264/AVC is called Multi-view Video Coding (MVC) to efficiently represent video signals simultaneously acquired by multiple cameras. MVC aims at applications such as 3D Television and Free Viewpoint Video. These new applications enable new user experiences including stereoscopic and head motion parallax viewing as well as free viewpoint navigation through scenes. Also alternative techniques such as single/multi-view plus depth and computer graphics compression are analyzed. Another extension of H.264/AVC is scalable video coding. Modern video transmission systems using the Internet and mobile networks are typically characterized by a wide range of connection qualities and receiving devices. Scalable video coding providing adaptation to error rates, throughput, power resources and spatial formats is a highly attractive option for modern video transmission applications as will be shown. These functionalities provide enhancements to transmission applications such as video streaming over 3GPP mobile, ad-hoc and peer-to-peer networks, mobile TV, as well as video conferencing.

Thomas Wiegand is the head of the Image Communication Group in the Image Processing Department of the Fraunhofer Institute for Telecommunications - Heinrich Hertz Institute Berlin, Germany. He received the Dipl.-Ing. degree in Electrical Engineering from the Technical University of Hamburg-Harburg, Germany, in 1995 and the Dr.-Ing. degree from the University of Erlangen-Nuremberg, Germany, in 2000. His research interest include video processing and coding, multimedia transmission, semantic image representation, as well as computer vision and graphics.

From 1993 to 1994, he was a Visiting Researcher at Kobe University, Japan. In 1995, he was a Visiting Scholar at the University of California at Santa Barbara, USA. From 1997 to 1998, he was a Visiting Researcher at Stanford University, USA and served as a consultant to 8x8, Inc., Santa Clara, CA, USA. He is currently a member of the technical advisory boards of the two start-up companies Layered Media, Inc., Rochelle Park, NJ, USA and Stream Processors, Inc., Sunnyvale, CA, USA.

Since 1995, he is an active participant in standardization for multimedia with successful submissions to ITU-T VCEG, ISO/IEC MPEG, 3GPP, DVB, and IETF. In October 2000, he was appointed as the Associated Rapporteur of ITU-T VCEG. In December 2001, he was appointed as the Associated Rapporteur / Co-Chair of the JVT. In February 2002, he was appointed as the Editor of the H.264/AVC video coding standard and its extensions (FRExt and SVC). In January 2005, he was appointed as Associated Chair of MPEG Video.

In 1998, he received the SPIE VCIP Best Student Paper Award. In 2004, he received the Fraunhofer Award for outstanding scientific achievements in solving application related problems and the ITG Award of the German Society for Information Technology. Since January 2006, he is an Associate Editor of IEEE Transactions on Circuits and Systems for Video Technology.

3-D DISPLAYS AND SIGNAL PROCESSING: AN ANSWER TO 3-D ILLS?

Janusz Konrad Boston University, USA

Three-dimensional perception of our surroundings is a natural part of daily life. Over the last 100+ years many attempts have been made to replicate the 3-D experience by means of various photo-, film- or print-based systems. While some have achieved limited commercial success, none have attained equal status to their 2-D counterparts. However, recent advances in electronic display technologies, digital signal processing techniques and computer graphics tools, promise a new era for 3-D displays. In fact, signal processing algorithms tuned to the unique problems of 3-D imaging will likely be the enabling technology for the emerging 3-D display systems. In this talk. I will discuss the major 3-D display technologies in use today, from simple glasses-based systems (colored, polarized, shuttered), through glasses-free parallax-barrier and microlens displays, to holographic, and volumetric display devices. I will describe the underlying physics as well as the associated benefits and deficiencies. I will emphasize the common principles that all 3-D displays share, and discuss issues such as sampling, multiplexing and rendering, particularly critical to dynamic 3-D display systems. I will highlight the role signal processing has played in addressing deficiencies of some 3-D displays, and will also point out the still unsolved problems awaiting signal processing solutions.

Janusz Konrad received M.Eng.degree from the Technical University of Szczecin, Poland in 1980, and the Ph.D. degree from McGill University, Montréal, Canada, in 1989. From 1989 to 2000 he was with INRS-Télécommunications (University of Québec), Montréal. Since 2000 he has been an Associate Professor at the Department of Electrical and Computer Engineering, Boston University. In the past, he collaborated with Imax Corporation, Bell-Northern Research, Digital Equipment Corp., and EMC Corp. He was an Associate Editor for the IEEE Transactions on Image Processing and IEEE Signal Processing Letters, as well as Technical Program Co-Chair for the IEEE International Conference on Image Processing (ICIP-2000), and Tutorials Co-Chair for the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP-2004), and member of the Image and Multidimensional Signal Processing Technical Committee of the IEEE Signal Processing Society. Currently, he is an Associate Technical Editor for IEEE Communications Magazine, Associate Editor for EURASIP Journal on Image and Video Processing, and a program committee member of several international conferences and workshops. He is a co-recipient, jointly with Dr. N. Božinović, of the 2004-2005 EURASIP Image Communication Best Paper Award and the IEEE 2001 Signal Processing Magazine award for a paper co-authored with Dr. C. Stiller. His interests are in the areas of image and video compression and processing, stereoscopic and 3-D imaging, multidimensional signal processing and computer vision.

EURASIP Fellow Inaugural Lectures

The EURASIP Administrative Committee (AdCom) has recently decided to initiate a 'Fellowship

Programme', to recognize outstanding achievements of its members and volunteers. Each year, a select

group of signal processing researchers will be elevated to 'EURASIP Fellow', the Association's now

most prestigious honor.

It is my pleasure to announce the result of the 2007 `EURASIP Fellows’ selection process. In

recognition of their many important contributions to the field of signal processing, the EURASIP

AdCom elevates the following individuals to `EURASIP Fellow 2007’:

Prof. Peter Grant, University of Edinburgh, Edinburgh, UK

Prof. Wolfgang Mecklenbräuker, Vienna University of Technology, Vienna, Austria

Prof. Peter Stoica, Uppsala University, Uppsala Sweden

Prof. Martin Vetterli, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland

I wish to congratulate our new EURASIP Fellows on this most important achievement. The Fellowship

Awards will be presented at the conference. In addition, to highlight the new Fellowship Programme,

Prof. Peter Grant and Prof. Martin Vetterli will present a 'Fellow Inaugural Lecture'.

Marc Moonen

President EURASIP

REVIEV OF REAL WORLD MIMO SYSTEM PERFORMANCE

Peter Grant

University of Edinburgh, United Kingdom

This review lecture will introduce the basic techniques which are adopted in multiple input multiple output (MIMO) systems and show the potential theoretical increase in data throughput or transmission efficiency in bit/s/Hz. This is dependent on a rich multipath environment supporting many unique propagation modes. Modelling and actual results will be presented from the universities of Bristol and Illmeneau on typical channel characteristics and provide practically realisable channel capacity in urban LOS and non-LOS channels. The presentation will also include real time hardware implementations of the receiver architectures from TU Vienna which use FPGA solutions. These provide particularly rapid design time. It will also describe simplified receiver designs using sphere decoders which achieve close to maximum likelihood performance combined with rapid FPGA implementation plus a high data rate capability.

Peter Grant, born in St. Andrews, received the B.Sc. degree in electronic engineering from the Heriot-Watt University, in 1966, the Ph.D. degree from the University of Edinburgh, in 1975, and an honorary DEng from the Heriot-Watt in 2006. He worked intially in radiocommunications for the Plessey Company, before he was appointed to a research fellowship at the University of Edinburgh. He was subsequently promoted to a Professor of Electronic Signal Processing in 1987 and in 2002 appointed head of the School of Engineering and Electronics.

During academic year 1977-78, he was a visiting professor at the Ginzton Laboratory, Stanford University, and in 1985-86 he was a visiting staff member at the MIT Lincoln Laboratory. He was awarded the 82nd (2004) Faraday Medal by the Institution of Electrical Engineers (IEE) for his work on CDMA receiver designs and adaptive filters. In 1974 and again in 1977 he was awarded the Bulgin premium from the then Institution of Electronic and Radio Engineers and in 1982 their Lord Mountbatten premium. In 1994 he was awarded the IEE Marconi and Langham Thompson premia.

Professor Grant was president of EURASIP, the European Association for Signal Processing, in 2000-2002, chairman of EUSIPCO-94 and technical programme chairman for ICASSP-89 international conferences. In 1998 he was appointed by the the US IEEE Signal Processing Society as a distinguished lecturer on DSP for Mobile Communications, presenting at 25 locations over five continents.

He served from 1980-1996 as an honorary editor of IEE Proceedings title "Vision Image and Signal Processing". He was chair of the 2001 Universities Funding Council research assessment panel for the UK Electrical Engineering Departments and has served as research assessor at: Queensland University of Technology, University of West Australia, City University Hong Kong and ETHz in Zurich. He is a member of the Scottish Science Advisory Committee.

He holds fellowships of the IEEE, IEE, Royal Academy of Engineering and the Royal Society of Edinburgh.

SAMPLING SPARSE SIGNALS AT OCCAM'S RATE

Martin Vetterli Ecole Polytechnique Fédérale de Lausanne, Switzerland and University of California, Berkeley, USA

Joint work with T. Blu, L. Coulot, A. Hormati and P.L.Dragotti. Shannon's sampling theorem gives a sufficient condition for reconstructing the projection of a signal onto the subspace of bandlimited functions, and this by taking inner products with a sinc function and its shifts. Recently, this framework has been extended to classes of non-bandlimited signals and their perfect reconstruction from a suitable projection. This gives a sharp result on the sampling and reconstruction of sparse continuous- time signals, namely that 2K measurements are necessary and sufficient to perfectly reconstruct a K-sparse continuous-time signal. We first review this result and show that it relies on structured Vandermonde measurement matrices, of which the Fourier matrix is a particular case. Because of this structure, fast, O(K^3) methods exist. When then generalize these results to a number of cases where sparsity is present, including piecewise polynomials as well as local measurement kernels like splines. Of course, real cases always involve noise, and thus, retrieval of sparse signals in noise is considered. Lower bounds by Cramer-Rao are given, and an iterative algorithm due to Cadzow is shown to perform close to optimal over a wide range of signal to noise ratios. This indicates the robustness of such methods. Next, we consider the connection to compressive sampling, a recent approach involving random measurement matrices, a discrete set up, and retrieval based on convex optimization. We compared the two approaches, highlighting differences, similarities, and respective advantages.

Martin Vetterli received his Engineering degree from ETH in Zurich, his MS from Stanford and his Ph.D. from EPFL in Lausanne. I n 1986, he joined Columbia University in New York, first with the Center for Telecommunications Research and then with the Department of Electrical Engineering where he was an Associate Professor of Electrical Engineering. In 1993, he joined the University of California at Berkeley, were he was Full Professor until 1997. Since 1995, he has been a Professor at EPFL, where he headed the Communication Systems Division (1996/1997) and heads the Audiovisual Communications Laboratory. From 2001 to 2004 he directed the National Competence Center in Research on mobile information and communication systems. He has also been Vice-President for International Affairs at EPFL since October 2004. His research interests are in the areas of applied mathematics, signal processing, and communications.

Tutorials

WHAT ENERGY TO MINIMIZE?

Mila Nikolova

Centre de Mathématiques et de Leurs Applications (CMLA), ENS de Cachan, France Many applications in signal and image processing are solved by minimizing an energy function composed of a data-fidelity term and a regularization term. Such energies are classically defined either from a PDE standpoint or in a Bayesian MAP estimation framework. In both approaches, the control on the solutions remains limited. Finer knowledge of these energies and of their minimizers is the key to obtain good solutions. This tutorial will present a systematic approach to the problem of the choice of pertinent energies for signal and image reconstruction. The last decade an important effort was done in order to gain an insight into these energies and the solutions they yield. Behind the wide variety of this research, several important orientations can be drawn. The interplay between energy minimization and constrained minimization gave rise to interpretations of the parameters and opened connections with the models based on frames (e.g. splines, wavelets or packets). Studying the spaces of functions (e.g. signals and images) underlying different energies helped to better understand sampling and modeling, and gave rise to new tools for reconstruction and decomposition. It also launched an examination and a clarification of the goals of signal and image reconstruction. Challenging theories established bridges between disparate methods based on energy minimization, diffusion and shrinkage estimation of frame representations. Even though holding in restricted conditions, these results suggest various practical ramifications for reconstruction and optimization. We focus more closely on the energy minimizers since a solution, defined as the minimizer of an energy, is an implicit function of both the data and the shape of the energy. This point of view raises the question of how the features of the reconstructed signals and images are determined by the shape of the energy. It hence provides a framework to unify the theory on energy minimization methods and to address rigorously the problem of the choice of energies for signal and image reconstruction. The objectives of this tutorial are the following:

• to understand the practical issues governing the proper choice of an energy for signal and image reconstruction; • to show how to conceive energies in such a way that their minimizers exhibit some desired properties; • to provide a systematic way to compare existing energies for signal and image reconstruction.

This talk is based on a series of analytical results which characterize some essential features exhibited by the minimizers of regularized energies, in connection with the shape of the energy. Points of interest are for instance the recovery of homogeneous regions, textures and edges, and the processing of signals and images containing outliers or spikes. These are shown to be determined by some attributes of the energy function relevant to its (non)smoothness or its (non)convexity. Numerical examples are used to illustrate the theory and stability results are provided. Afterwards we present several applications where specific energies are conceived using the mathematical results on minimizers. Indications on implementation issues are also given. The minimizer approach and results invoke a critical analysis of the ways to construct energies and a new understanding of modeling. By way of conclusion, open questions ranging from concepts to practical signal and image reconstruction are discussed.

Mila Nikolova received the Ph.D. degree in signal processing from the Université Paris-Sud, Paris, France, in 1995. Currently, she is senior research fellow with the National Center for Scientific Research (CNRS), France and performs her research at the Centre de Mathématiques et de Leurs Applications (CMLA), ENS de Cachan, France. Her research interests are in Image and signal reconstruction, Regularization and variational methods, Scientific computing. The last few years she published a series of papers analyzing the properties of the minimizers of regularized energy functions.

MULTIRATE SIGNAL PROCESSING IN COMMUNICATION SYSTEMS

Frederick Harris San Diego State University, San Diego, California, USA

Multirate Signal Processing offers the option of accomplishing a given digital signal processing task with the smallest expenditure of energy and with the smallest signal processing resources. Is everyone listening? This is an amazing statement and consequently, we should all be advocates of multirate signal processing! Multirate signal processing has found its way into many DSP text books and is part of the title of at least six textbooks. The most common theme in these many books and in these many chapters is the critically sampled perfect reconstruction filter bank. These amazing systems, embedded in all MP3 recorders, enable aliasing at the filter bank band edges due to down sampling to be canceled during the up sampling in the reconstruction process. What we learn here is that aliasing is OK if you do it right! What we miss is how aliasing can be used in a very much wider class of tasks, and in particular to numerous communication system applications. These include timing recovery in digital receivers, arbitrary resamplers in modulators and demodulators, clock domain alignment between arbitrary asynchronous systems, cascade down-sampling and up-sampling to obtain minimum resource filtering, alias based up or down conversion embedded in the down-sampling and up-sampling process, and many others. This tutorial will review multirate filters in the standard FIR configuration as well as the IIR configurations. We will emphasis concept, will be light on mathematics, and will develop and illustrate many communication system applications.

Frederick J Harris holds the CUBIC Signal Processing Chair of the Communication Systems and Signal Processing Institute at San Diego State University where since 1967 he has taught courses in Digital Signal Processing and Communication Systems. He has extensive practical experience in communication systems, high performance modems, sonar and advanced radar systems and high performance laboratory instrumentation. He holds a number of patents on digital receiver and DSP technology and lectures throughout the world on DSP applications. He consults for organizations requiring high performance, cost effective DSP solutions.

His special areas of concentration are Signal Processing Algorithms, and in particular Multirate Signal Processing, Modem Design, Synchronization Techniques, and Fast Algorithms. He is the author of the book "Multirate Signal Processing for Communication Systems" and has contributed to a number of other books and encyclopedia articles on various DSP techniques. In 1990 and 1991 he was the Technical and then the General Chair of the Asilomar Conference on Signals, Systems, and Computers and was Technical Chair of the 2003 Software Defined Radio Conference and the Technical Chair of the 2006 Wireless Personal Multimedia Conference. He became a Fellow of the IEEE in 2003, cited for contributions of DSP to communications systems and has been awarded the 2006 Lifetime Achievement award by the Software Defined Radio Forum.

RECENT ADVANCES IN VIDEO CODING

Thomas Wiegand Fraunhofer Institute for Telecommunications, Berlin, Germany

With the introduction of H.264/AVC, significant improvements have recently been demonstrated in video compression capability. These advances have been brought about by improvements in signal processing techniques along with a relaxation of the bounds on practical computing power. While H.264/AVC has found its way into numerous transmission and storage applications spanning the range from smallest (Mobiles, iPod) to largest video resolutions (HDTV, Blu-Ray, HD DVD), new techniques are being developed and standardized. One new extension of H.264/AVC is called Scalable Video Coding (SVC) allowing partial transmission and decoding of a bitstream resulting in lower temporal or spatial resolutions or reduced quality. SVC provides functionalities such as graceful degradation in lossy transmission environments as well as bit rate, format, and power adaptation. These functionalities provide enhancements to transmission applications such as video streaming over 3GPP mobile, ad-hoc and peer-to-peer networks as well as video conferencing. Another new extension of H.264/AVC is called Multi-view Video Coding (MVC) to efficiently represent video signals simultaneously acquired by multiple cameras. MVC aims at applications such as 3D Television and Free Viewpoint Video. These new applications enable new user experiences including stereoscopic and head motion parallax viewing as well as free viewpoint navigation through scenes. The tutorial will cover the following: - H.264/AVC - History - Technology - Performance - Profiles - SVC - History - Technolgy - Performance - Video transmission with SVC - Profiles - MVC - History - Technology - Potential applications

Thomas Wiegand is the head of the Image Communication Group in the Image Processing Department of the Fraunhofer Institute for Telecommunications - Heinrich Hertz Institute Berlin, Germany. He received the Dipl.-Ing. degree in Electrical Engineering from the Technical University of Hamburg-Harburg, Germany, in 1995 and the Dr.-Ing. degree from the University of Erlangen-Nuremberg, Germany, in 2000. His research interest include video processing and coding, multimedia transmission, semantic image representation, as well as computer vision and graphics.

From 1993 to 1994, he was a Visiting Researcher at Kobe University, Japan. In 1995, he was a Visiting Scholar at the University of California at Santa Barbara, USA. From 1997 to 1998, he was a Visiting Researcher at Stanford University, USA and served as a consultant to 8x8, Inc., Santa Clara, CA, USA. He is currently a member of the technical advisory boards of the two start-up companies Layered Media, Inc., Rochelle Park, NJ, USA and Stream Processors, Inc., Sunnyvale, CA, USA.

Since 1995, he is an active participant in standardization for multimedia with successful submissions to ITU-T VCEG, ISO/IEC MPEG, 3GPP, DVB, and IETF. In October 2000, he was appointed as the Associated Rapporteur of ITU-T VCEG. In December 2001, he was appointed as the Associated Rapporteur / Co-Chair of the JVT. In February 2002, he was appointed as the Editor of the H.264/AVC video coding standard and its extensions (FRExt and SVC). In January 2005, he was appointed as Associated Chair of MPEG Video.

In 1998, he received the SPIE VCIP Best Student Paper Award. In 2004, he received the Fraunhofer Award for outstanding scientific achievements in solving application related problems and the ITG Award of the German Society for Information Technology. Since January 2006, he is an Associate Editor of IEEE Transactions on Circuits and Systems for Video Technology.

DEFORMABLE MODELS IN MEDICAL IMAGE PROCESSING; ADVANCES IN IMAGE GUIDED RADIOTHERAPY

Bogdan J. Matuszewski*, Christopher J. Moore** * Applied Digital Signal and Image Processing Research Centre Department of Technology, University of Central

Lancashire, United Kingdom ** Christie Hospital NHS Trust, North Western Medical Physics , United Kingdom

The tutorial will describe the state-of-the-art in modelling deformations with applications in medical image processing. The focus of the tutorial will be on the use of deformable techniques in medical image registration, but other areas including segmentation will be also discussed. Image registration and segmentation are key enabling technologies in the medical image processing which form a part of almost any medical image processing algorithm. They are essential to provide common reference frame, to acquire measurements enabling robot assisted interventions, post-surgical assessment, monitoring of disease progress, detection of abnormalities, measurement of radiotherapy radiation dose, or the functional imaging and its mapping to the anatomical structures. Although different applications areas of deformable models in the medical image processing will be examined the main emphasis will be on their use in radiation therapy. Radiation therapy is designed to cure localised cancer by repeatedly targeting a tumour with small doses of high energy radiation over many days. The treatment plan provides a sophisticated estimate of the radiation dose distribution inside the patient, usually as they appear on a computer tomography (CT) scan before treatment. However, some time later, when treatment is in progress and radiation is being directed at the tumour, the options for monitoring the patient's internal anatomy are very limited. The tutorial will draw from new research endeavour undertaken by the authors on digitally 'seeing and measuring' what is happening to the patient during their radiation therapy. Their research project, named Metrology Guided Radiation Therapy (MEGURATH), is developing technologies and methodologies for live measurement of a patient's position and shape, linking this directly to internal anatomy during treatment using optical surface scanner and on-board imager (Image Guided Radiation Therapy) The tutorial will introduce taxonomy of the deformable methods. The special attention will be put on explaining the underling methodology of using deformable modelling. This will be reinforced by the structure of the tutorial, where key design choices, namely similarity measures, displacement/shape models, and optimisations procedures, will be identified. The main implementation aspects of deformable models will be clearly specified with discussion of different options available for each step, their performance and examples of their use. Throughout the tutorial the ideas introduced will be visualised using practical cases with real multimodality medical imaging data including radiotherapy treatment planning CT scans (RTPCT), cone beam CT scans (CBCT), dynamic magnetic resonance scans (dMR) and optical body surface scanning (OBSS) data. In contrast to rigid/affine registration, the quantitative assessment of the accuracy of deformable registration is a challenging problem. It has not yet been fully solved. Number of different techniques have been proposed for this including use of human body phantoms and simulation of plausible deformations using biophysical modelling techniques. A short overview of these techniques will be included in the tutorial for completeness. The tutorial will conclude with a few case studies to stress overall methodology and to show a very practical nature of deformable models applied in medical imaging. The tutorial is designed to give an extensive introduction to deformable models, their use, and to stimulate interest in this field. Some of the methods will be explained in more details to give a better taste of their complexity and to satisfy more accomplished participants. No priori knowledge of the medical imaging techniques will be required and although majority of the examples will be drawn from the medical domain the described techniques have much wider applications. The structure of the tutorial is as follow:

• Introduction o Deformable models versus global rigid/affine models. o Registration, segmentation, fusion and other uses of the deformable models in medical image

processing. o Medical imaging modalities and their properties with respect to registration and segmentation.

• Similarity Measures o Sum of square differences and correlations measures o Mutual information, normalised mutual information, and correlation ratio o Explicit modelling of the image intensity

• Deformable Models o Feature versus intensity based methods o Free form deformations o Constraints imposed on deformable models

o Elastic, plastic, fluid and diffused models o Variational models o Stochastic models, stochastic shape models and stochastic displacement models, Bayesian

methodology, likelihood function, estimation of the priors. o Spatio-temporal models

• Optimisation Procedures o Numerical optimisation (simplex, steepest descent, conjugate gradients, LM, …) o Direct and indirect methods in solving variational problems o Finite element method (FEM) versus finite difference method (FDM) o EM, Monte Carlo and sequential Monte Carlo methods

• Validation Methods • Case Studies

o Spherical harmonics PCA bladder segmentation. o Deformable registration between cone-beam CT and radiotherapy planning CT for treatment planning

grey scale calibration o Dynamical statistical shape priors for level set based organ segmentation and deformable registration.

Bogdan Matuszewski received MSc (with honors), and PhD degrees, both in electronic engineering from Wrocław University of Technology (Poland). Currently he is a senior lecturer in the Department of Technology, and a Head of the Robotics and Vision Laboratory at the University of Central Lancashire, United Kingdom. He is a Member of the IEEE and BMVA. Dr Matuszewski has published over 50 research papers in different areas of computer vision and image processing. His recent research interests include use of Bayesian methodology for modelling, classification and tracking; deformable models and their applications to data registration and segmentation; estimation of a human posture; and image based rendering. He collaborated with many industrial partners including BAE Systems, Alenia Aerospazio and EADS. Most recently he has been working with number of universities from the Atlantic Arch region on hyper-spectral imaging project PIMHAI; with the computer vision group at the Heriot-Watt University on multi-view representation for view synthesis; and with Christie Hospital and Liverpool John Moores University on Metrology Guided Radiation Therapy project.

Prof. Christopher J. Moore, North Western Medical Physics, Christie Hospital, Manchester M20 4BX Gained a 1st class honours degree in Physics from Manchester University in 1976 and joined North Western Medical Physics at the Christie Hospital, where he now leads the Developing Technologies Section of Radiotherapy Physics. He is a Chartered Scientist and State Registered Clinical Scientist. His M.Sc is in Computational Physics, which he obtained from Salford University in 1982. He obtained his Ph.D in Image Analysis from Manchester University in 1988. He holds a visiting chair in Medical Physics at Liverpool John Moores University. He has been responsible for research in clinical-signal analysis, diagnostic and radiotherapy image processing, image-guided irradiation of tumours, dosimetric and endocrine modelling, and radiobiological prediction. He developed clinical facilities for CT-image assisted planning of cervical cancer therapy and led the creation of image and graphics based computerised conformal radiotherapy using the Western world's first multi-leaf collimator for shaping mega-voltage X-ray treatment beams. Some 10,000 patients were treated with deliverables from these extended programmes. He has led or participated in nine UK and European research collaborations in the past decade, is an EU Expert scientific evaluator and a reviewer for the Engineering and Physical Science Research Council. He has over 100 peer reviewed publications.

IMAGE FUSION - PRINCIPLES, METHODS, AND APPLICATIONS

Jan Flusser, Filip Šroubek, Barbara Zitová Institute of Information Theory and Automation, Academy of Sciences of the Czech Republic

The term Image Fusion (IF) means in general an approach to extraction of information spontaneously adopted in several domains. The goal of image fusion is to integrate complementary multisensor, multitemporal and/or multiview information into one new image containing information the quality of which cannot be achieved otherwise. The term "quality" depends on the application requirements. Image fusion has been used in many application areas. In remote sensing and in astronomy, multisensor fusion is used to achieve high spatial and spectral resolution by combining images from two sensors, one of which has high spatial resolution and the other one high spectral resolution. Numerous fusion applications have appeared in medical imaging, like simultaneous evaluation of CT (computer tomography), NMR (nuclear magnetic resonance) and/or PET (positron emission tomography) images to obtain more complete information about the patient, and in military applications (combining visible and infrared or radar data for target localization and missile navigation). In the case of multiview fusion, a set of images of the same scene taken by the same sensor but from different viewpoints is fused to obtain an image with higher resolution than the sensor normally provides or to recover the 3D representation of the scene (shape from stereo). The multitemporal approach recognizes two different aims. Images of the same scene are acquired at different time instances either to find and evaluate changes in the scene or to obtain a less degraded image of the scene. The former aim is common in medical imaging, especially in change detection of organs and tumors, and in remote sensing for monitoring land or forest exploitation. The acquisition period is usually months or years. The latter aim requires the different measurements to be much closer to each other, typically in the scale of seconds, and possibly under different conditions. Recent development of the field has proved that IF can be also a useful tool for resolution enhancement. The list of applications mentioned above illustrates the diversity of problems we face when fusing images. It is impossible to design a universal method applicable to all image fusion tasks. Every method should take into account not only the fusion purpose and the characteristics of individual sensors, but also particular imaging conditions, imaging geometry, noise corruption, required accuracy and application-dependent data properties. In this tutorial, we categorize the IF methods according to the data entering the fusion and according to the fusion purpose. We distinguish the following categories: 1) Multiview fusion of images from the same modality and taken at the same time but from different viewpoints. 2) Multimodal fusion of images coming from different sensors (visible and infrared or CT and NMR). 3) Multifocus fusion of images of 3D scene taken repeatedly with various focal length. 4) High res - Low res fusion of two images where the first one has high spectral but low spatial resolution and vice versa. 5) Multitemporal fusion of images taken at different times in order to detect changes between them. 6) Fusion for image restoration. Fusion two or more images of the same scene and modality, each of them blurred and noisy, may lead to a deblurred and denoised image. Multichannel deconvolution is a typical representative of this category. 7) Superresolution fusion of two or more images of the same scene and modality, all having low spatial resolution, may provide us a high-resolution image. 8) Fusion for synthesis. By fusing two or more real images it is possible to create artificial images of objects which never existed or realistic images of objects which exist but were not photographed in a desired time/stage. In each category, fusion consists of two basic steps: image registration, which brings the input images to spatial alignment, and combining the image functions (intensities, colors, etc). We present a survey of traditional and up-to-date fusion methods and demonstrate their performance by practical experiments from various application areas. Special attention is paid to Fusion for image restoration and to Superresolution fusion, because these two groups are extremely important for producers and users of low-resolution imaging devices such as mobile phones, camcorders, web cameras, and security and surveillance cameras. We propose a unifying system that simultaneously estimates blurs and recovers the original undistorted image, all in high resolution, without any prior knowledge of the blurs and original image. We accomplish this by formulating the problem as constrained least squares energy minimization with appropriate regularization terms, which guarantee close-to-perfect solution. We demonstrate the performance of the superresolution fusion on many examples, namely on car licence plate recognition and face recognition. Live demo showing the fusion of webcam images will run on a laptop during the tutorial.

Jan Flusser received the M.Sc. degree in mathematical engineering from the Czech Technical University, Prague, Czech Republic in 1985 and the Ph.D. degree in computer science from the Czechoslovak Academy of Sciences in 1990. Since 1985 he has been with the Institute of Information Theory and Automation, Academy of Sciences of the Czech Republic, Prague. Since 1995 he has been holding the position of a head of Department of Image Processing. Since 1991 he has been also affiliated with the Faculty of Mathematics and Physics, Charles University, Prague and with the Czech Technical University, Prague (full professorship in 2004), where he gives undergraduate and graduate courses on Digital Image Processing and Pattern Recognition, and specialized graduate course on moment invariants and wavelets. Jan Flusser has a 20-years experience in basic and applied research on invariant-based pattern recognition. He has been involved in applications in remote sensing, medicine, and astronomy. He has authored and coauthored more than 100 research publications in these areas. Some of his journal papers became classical and are frequently cited. Filip Šroubek received the M.Sc. degree in computer science from the Czech Technical University, Prague, Czech Republic in 1998 and the Ph.D. degree in computer science from the Charles University, Prague, Czech Republic in 2003. From 2004 to 2006, he was on a postdoctoral position in the Instituto de Optica, CSIC, Madrid, Spain. He is currently with the Institute of Information Theory and Automation and partially also with the Institute of Radio Engineering and Electronics, where both institutes are part of the Academy of Sciences of the Czech Republic. Filip Sroubek is an author of two book chapters and over 25 journal and conference papers on image fusion, blind deconvolution, super-resolution, and related topics. Barbara Zitová received the M.Sc. degree in computer science from the Charles University, Prague, Czech Republic in 1995 and the Ph.D. degree in computer science from the Charles University, Prague, Czech Republic in 2000. Since 1995 she has been with the Institute of Information Theory and Automation, Academy of Sciences of the Czech Republic, Prague. She also gives tutorials on Image Processing and Pattern Recognition at the Czech Technical University. Jointly with J. Flusser, she gives specialized graduate course on moment invariants and wavelets. Barbara Zitova has a 10-years experience in image analysis. She is an author of a book chapter in Invariants for Pattern Recognition and Classification (M.A. Rodrigues ed., World Scientific, 2000) and of 20 journal and conference papers on moment invariants and related topics. Her paper "Image Registration Methods: A Survey" (Image and Vision Computing, vol. 21, pp. 977-1000, 2003) has became a major reference in image registration.

SPATIALLY ADAPTIVE LOCAL APPROXIMATIONS IN SIGNAL AND IMAGE

PROCESSING: VARYING-SCALE POLYNOMIALS, ANISOTROPIC ADAPTATION, SHAPE-ADAPTIVE TRANSFORMS

Alessandro Foi, Vladimir Katkovnik, Karen Egiazarian Institute of Signal Processing, Tampere University of Technology, Finland

We present an overview on modern adaptive solutions for signal reconstruction problems based mainly on combining two independent nonparametric estimation ideas: the local polynomial approximation (LPA) and the intersection of confidence intervals (ICI) rule. The LPA is a technique that is applied for nonparametric estimation using a polynomial data fit in a sliding window. The ICI rule is a criterion used for the adaptive selection of the size (scale) of this window. The resulting LPA-ICI estimators are nonlinear filters that are adaptive to the unknown smoothness of the signal. The local polynomial approximation is originated from an old idea known under different names: moving (sliding, windowed) least-square, Savitzky-Golay filter, moment filters, reproducing kernels, singular convolution kernels, etc. However, combined with the new adaptation technique it becomes a novel powerful tool. The window size, interpreted also as scale, is the key parameter of this technique. The terms "window size", "bandwidth", and "scale" are interchangeable here. The idea of the ICI scale-adaptation is as follows. The algorithm searches for a largest local vicinity of the point of estimation where the local polynomial approximation assumptions fit well to the data. The estimates are calculated for a number of different scales and compared. The adaptive scale is defined as the largest for which the estimate does not differ significantly from the estimates corresponding to the smaller scales. The ICI rule defines the adaptive scale for each point (pixel, voxel) of the signal. In this way, we arrive to a pointwise-adaptive signal and image processing. Asymptotically, these adaptive estimators allow to get a near-optimal quality of the signal recovery. The anisotropic implementation of the LPA-ICI, based on the use of multi-directional kernels, gives further improvement to the adaptivity of the method, providing an efficient tool especially for image denoising, differentiation and inverse-imaging problems. The latest development of these techniques goes beyond the traditional fixed-order polynomial models, replacing them with more general transforms defined on arbitrarily-shaped domains. This corresponds to multidimensional local polynomial and non-polynomial approximations with adaptive order and support. The efficient realization of these transforms is illustrated. We conclude with a short discussion on the ongoing transition from local techniques to non-local ones (e.g., non-local means, patch-based estimators, block-matching 3D filtering), which appears as the latest and very successful trend in image denoising. Discussion and comparison with other relevant approaches (including anisotropic diffusion, TV minimization, wavelets, curvelets, etc.) is given, highlighting the similarities and differences between the techniques. The tutorial is accompanied by numerous experimental examples where these methods are applied to competitive image processing problems. Applications include image denoising, deblurring (deconvolution), deblocking and deringing, gradient estimation, edge-detection, inverse-halftoning, and color image processing. Matlab software, which implements the presented techniques and experiments, is provided. Content: local polynomial approximation (theory and methods); linear smoothing and differentiation; adaptive scale selection: intersection of confidence intervals rule; adaptive algorithms; directional LPA; adaptive aggregation of multidimensional estimates, anisotropic LPA-ICI; spatially adaptive anisotropic regularization for deconvolution problems; image deblurring; shape-adaptive transforms, shape-adaptive DCT algorithms; non-local methods.

Alessandro Foi has received the M.Sc. and the Ph.D. degrees in mathematics from Universita degli Studi di Milano (Italy) in 2001 and from Politecnico di Milano in 2005, respectively. His research interests include mathematical and statistical methods for signal processing, functional analysis, and harmonic analysis. Currently, he is a researcher at the Institute of Signal Processing, Tampere University of Technology (Finland). His work focuses on spatially adaptive algorithms for denoising and deblurring of digital images and on noise modeling for digital imaging sensors. Vladimir Katkovnik received the M.Sc., Ph.D., and D.Sc. degrees in technical cybernetics from the Leningrad Polytechnic Institute, Leningrad, Russia, in 1960, 1964, and 1974, respectively. From 1964 to 1991, he held the positions of Associate Professor and Professor at the Department of Mechanics and Control Processes, Leningrad Polytechnic Institute. From 1991 to 1999, he was a Professor of statistics with the Department of the University of South Africa, Pretoria. From 2001 to 2003, he was a Professor of mechatronics with the Kwangju Institute of Science and Technology, Korea. From 2000 to 2001 and since 2003 he is a Research Professor with the Institute of Signal Processing, Tampere University of Technology, Tampere, Finland. He has published seven books and more than 250 papers. His research interests include stochastic signal processing, linear and nonlinear filtering, nonparametric estimation, imaging, nonstationary systems, and time-frequency analysis. Karen Egiazarian was born in Yerevan, Armenia, in 1959. He received the M.Sc. degree in mathematics from Yerevan State University in 1981, the Ph.D. degree in physics and mathematics from Moscow State University, Moscow, Russia, in 1986, and the D.Tech. degree from the Tampere University of Technology (TUT), Tampere, Finland, in 1994. He has been Senior Researcher with the Department of Digital Signal Processing, Institute of Information Problems and Automation, National Academy of Sciences of Armenia. Since 1996, he has been an Assistant Professor with the Institute of Signal Processing, TUT, where he is currently a Professor, leading the Transforms and Spectral Methods group. His research interests are in the areas of applied mathematics, signal processing, and digital logic.

POWER LINE COMMUNICATIONS

Stefano Galli Panasonic

There is today growing interest in the prospects of re-using the existing power line infrastructure to provide broadband access to the home and LAN connectivity within the home or office. Besides the traditional access and home-LAN applications, power line communications (PLCs) also have other interesting applications. Today in the construction of vehicles, ranging from automobiles to ships, from aircraft to space vehicles, there is a power distribution system based on metallic conductors and this power distribution network may well perform double-duty, as an infrastructure supporting both power delivery and broadband digital connectivity. PLCs could also allow power utility companies to manage more efficiently their grid as well as to introduce novel post meter applications. Considerable effort has been devoted recently to determining accurate channel models for the power line environment. Although there are today two main approaches to channel modeling (multipath and transmission-line models), there is no widely accepted channel model similar to those derived for mobile radio or telephone channels. To complicate things even further, particular wiring and grounding practices in residential and commercial buildings often make channel modeling a very challenging task. The consequence of the lack of a universally agreed upon channel model is that a solid communications and information theoretic approach to power line communications is still missing. The tutorial will start by reviewing the major applications of PLCs (access, in-home, in-vehicle, smart grid, etc.), and pointing out world trends and market status. Moreover, a brief overview of standardization and PLC industry associations will be given. The course will continue addressing the existing channel models, with particular reference to recent results. Some inefficiencies and inaccuracies in capturing the physics of signal propagation, when particular topologies or particular wiring and grounding practices are taken into account, will also be pointed out. Finally, the suitability of modulation and coding techniques for the power line channel will be reviewed. The topic of PLCs is a very interdisciplinary one so it is impossible to give an in-depth course on it in a half-day tutorial. The course, however, will give a rather complete overview of the major issues related to power line communications (PLCs). The contents of the proposed tutorial are structured as follows: 1) Broadband applications today: access, in-home, in-vehicle, and beyond 2) Overview of IEEE standardization efforts and PLC industry associations 3) Main characteristics of power supply systems and of indoor wiring 4) The power line channel and its classical models

a. Transfer function b. Noise

5) Recent results on the modeling of the indoor power line channel transfer function a. Multi-conductor Transmission Line (MTL) theory approach b. From MTL to the transfer function c. Example of computation of a transfer function

6) Towards the definition of an "average" channel 7) Modulation & Coding for the power line channel

Stefano Galli received his M.S. degree and Ph.D. in Electrical Engineering from the University of Rome "La Sapienza" (Italy) in 1994 and 1998, respectively. He is currently a Lead Scientist in the Strategic R&D Planning Office of Panasonic in San Jose', California. His principal role is to research and develop techniques for Panasonic's next generation power line modems. He is also active in the standardization of PHY and MAC of power line communications within the IEEE P1901 Working Group, where he is co-Leader of the "Theoretical and Mathematical Models" Group, and co-author of Chapter 3 of the Informative Annex: "Theoretical/Mathematical Channel Models for BPL Systems." Prior to this position, Dr. Galli was a Senior Scientist in Bellcore (now Telcordia Technologies) from 1998 to 2006, where he worked in the Broadband Networking Research Department on several commercial and government projects. He also worked as a Teaching Assistant at the Info-Com Dpt. of the University of Rome from 1996 to 1998.

Dr. Galli is very active in the IEEE Communications Society (ComSoc). He is currently serving as Chair of the "Communications and Signal Processing" Technical Committee Cluster, reporting directly to the IEEE Vice President for Technical Activities, as Chair of the Technical Committee on "Power Line Communications", and he also serves on the On-Line Content Activity Board.

Dr. Galli has served as Technical Program Committee member in numerous conferences, has served as the General Co-Chair of the 2005 IEEE Workshop on Signal Processing Advances in Wireless Communications (SPAWC), as Vice-Chair of the General Symposium of the 2006 IEEE International Conference on Communications (ICC), as the Co-Chair of the General Symposium of the 2006 IEEE Global Communications Conference (Globecom), and he is currently serving as Chair of the Technical Program Committee of the IEEE International Symposium on Power Line Communications and its Applications (ISPLC), and of the 2007 IEEE International Conference on Access Networks (AccessNets). Dr. Galli also served as co-Guest Editor for the Feature Topic "Broadband is Power: Internet Access through the Power Line Network" (IEEE Communications Magazine, May 2003), and as the co-Guest Editor for the first IEEE Journal on Selected Areas in Communications (JSAC) special issue on Power Line Communications (July 2006). Dr. Galli is currently serving as Associate Editor for the IEEE Signal Processing Letters, Area Editor for Signal Processing for Communications.

Dr. Galli's research efforts are devoted to various aspects of xDSL systems, wireless/wired home networks, wireless communications, optical CDMA, and power line communications. His research interests also include detection and estimation, communications theory, and signal processing. He is an IEEE Senior Member, a reviewer for several IEEE journals and conferences, has published over 80 papers, and holds several international issued and pending patents.

DISTRIBUTED VIDEO CODING: BASICS, CODECS AND APPLICATIONS

Christine Guillemot and Aline Roumy AINRIA/IRISA, Campus Universitaire de Beaulieu, Rennes, France.

Distributed source coding has emerged as an enabling technology for sensor networks. It refers to the compression of correlated signals captured by different sensors which do not communicate between themselves. All the signals captured are compressed independently and transmitted to a central base station which has the capability to decode them jointly. Distributed source coding finds its foundation in the seminal work of Slepian-Wolf (1973) and Wyner-Ziv (1976). The minimum achievable rate for lossless compression of two statistically dependent memoryless sources is given by the joint entropy of the two sources. Slepian and Wolf have established that this lossless compression rate bound could be approached with a vanishing error probability for long sequences, even if the two sources are coded separately, provided that they are decoded jointly and that their correlation is known to both the encoder and the decoder. This theorem has led to a new coding paradigm known as distributed source coding. The lossy equivalent of the Slepian-Wolf theorem has been formulated later on by Wyner and Ziv. The proof of the Slepian-Wolf W theorem is based on random binning, which is non-constructive, i.e., it does not reveal how practical code design should be done. In 1974, Wyner suggested the use of parity check codes to approach the corner points of the Slepian-Wolf rate region. It is only recently that practical solutions based on channel capacity-achieving codes (block codes, turbo codes or LDPC codes) have been explored for applications ranging from video compression, resilient video transmission, to minimization of transmit energy in sensor networks. Video compression, as well as scalable video compression, has been recast into a distributed source coding framework leading to distributed video coding schemes targeting mainly low coding complexity and error resilience functionalities. Correlated samples (pixels or transform coefficients) from different frames are regarded as outputs of different sensors. However, the application of the Wyner-Ziv principles to video compression is not straightforward and requires solving a number of issues. The Distributed Source Coding principles apply quite naturally to the compression of video sequences captured of the same scene by several cameras. With respect to classical multiview coding techniques, DVC allows the exploitation of correlation between views without - or with limited - inter-sensor (that is inter-camera) communication. This tutorial will present the underlying theory as well as latest developments of distributed video compression for both monoview and multiview applications. Christine Guillemot is currently 'Directeur de Recherche' at INRIA, in charge of the TEMICS research group dealing with image modelling, processing, video communication and watermarking. She holds a PhD degree from ENST (Ecole Nationale Superieure des Telecommunications) Paris. From 1985 to October 1997, she has been with FRANCE TELECOM/CNET, where she has been involved in various projects in the domain of coding for TV, HDTV and multimedia applications, and co-ordinated a few (e.g. the European RACE-HAMLET project). From January 1990 to mid 1991, she has worked at Bellcore, NJ, USA, as a visiting scientist. Her research interests are signal and image processing, video coding, and joint source and channel coding for video transmission over the Internet and over wireless networks. She has served as Associate Editor for IEEE Trans. on Image Processing (2000-2003), and for IEEE Trans. on Circuits and Systems for Video Technology (2004-2006). She is a member of the IEEE IMDSP and of the IEEE MMSP technical committees. Aline Roumy received the Engineering degree from Ecole Nationale Supérieure de l'Electronique et de ses Applications (ENSEA), Cergy, France in 1996, the Master degree in June 1997 and the Ph.D. degree in September 2000 from the University of Cergy-Pontoise, France. During 2000-2001, she has been the recipient of a French Defense DGA/DRET postdoctoral fellowship and was a research associate at Princeton University, Princeton, NJ. On November 2001, she joined INRIA, Rennes, France. Her current research and study interests include the area of statistical signal processing, coding theory and information theory.

VISUAL MEDICINE: TECHNIQUES AND APPLICATIONS FOR COMPUTER-ASSISTED SURGERY

Dirk Bartz, Oliver Burgert University of Leipzig, ICCAS, Germany

One of the largest application domains of visual computing is medicine. Image processing methods are widely used to identify anatomical structures or to enhance the data quality of acquired image data. 3D computer graphics is employed to efficiently and effectively present complex anatomical and functional information of the human body. Surgeons and medical technicians interact with these data to explore access ways and to identify target areas and risk structures. All together, visual computing is essential in both daily health-care practice and in clinical research. In particular, recent developments in image acquisition, diagnostic procedures and minimally-invasive surgery require an advanced planning and intra-operative support through computer science methods. With the increased spatial and temporal resolution, 3D visualizations become important for selected diagnostic procedures and for a wide range of therapy planning scenarios ranging from radiation treatment planning to neurosurgical interventions. Finally, the advent of intraoperative navigation gives rise to augmented reality techniques to support medical doctors during surgery. In this tutorial, we will first give an introduction into medical imaging methods - such as data acquisition, data analysis, segmentation, registration and rendering - both in 2D and 3D. Based on this foundation, the course will further explore a variety of advanced topics of visual medicine. In particular, we will discuss special representation techniques that mimic minimally-invasive procedure, OR-fit mixed reality methods for surgery, and the analysis of the workflow for surgical procedures in the OR. Together, these topics form important components towards more realistic interaction with the virtualized human body. Besides the technical aspects, we will also discuss the advantages to traditional methods, but will also illustrate their specific and inherent limitations. Tutorial Syllabus: I Introduction, II Basics of Medical Imaging,

- Data Modalities, Volume Data, and Data Artifacts - Segmentation, Classification, Registration and Fusion - Visualization and Navigation

III Applications, - Intraoperative Navigation and Medical Mixed Reality - Virtual Endoscopy - Surgerical DICOM and Surgical Workflow Analysis

IV Discussion.

Dirk Bartz is Professor for Computer-Aided Surgery at the University of Leipzig. He was head of the research group on Visual Computing for Medicine of the University of TYbingen. He was member of the executive committee of the IEEE Visualization and Graphics Technical Committee (VGTC), speaker of the Eurographics Working Group on Parallel Graphics, and founding/steering member of the "Gesellschaft for Informatik" working groups on Medical Visualization and Data Visualization. In 1998, 2002, and 2004, he was co-chair of the "9th Eurographics Workshop on Visualization in Scientific Computing" (now known as EuroVis), and of the 4th and 5th "Eurographics Workshop on Parallel Graphics and Visualization".

Dirk Bartz studied computer science and medicine at the University of Erlangen-NYrnberg and at Stony Brook University. He received a Diploma (M.S.) in computer science from the University of Erlangen-Nurnberg in 1997, and in 2001 a Ph.D. in computer science and a habilitation in computer science from the University of Tubingen (all in Germany). His main research interests are in visual medicine and medical imaging, medical mixed reality, visualization of large datasets, man-machine interfaces, parallel/grid computing, and data visualization. In 2002, he received the NDI Young Investigator Award for his work in virtual endoscopy and intra-operative navigation.

Oliver Burgert is head of the research group "Scientific Methods" at the Innvation Center Computer Assisted Surgery (ICCAS) at the University Leipzig, Germany. He studied Informatics at Universitet Karlsruhe (TH), Germany. After that he worked as a research scientist in the "Collaborative Research Fund: Information Technology in Medicine" (SFB 414), and other medical simulation and planning projects in CMF-, heart-, and neurosurgery. He wrote his doctoral thesis on "Planning and Support of Shape Changing Surgical Interventions in Maxillo-facial Surgery using Volumetric Data" at the IAIM, Universitet Karlsruhe (TH) in 2005. His research is focused on description and analysis of surgical interventions (Surgical Workflows) and patient modelling. Standardization of patient model storage and communication in the OR in the context of DICOM are an application field of his research. He is involved in several DICOM project groups (Ontologies and Surgical Workflow, Image Processing and Display). Clinical application fields are simulation- and planning systems in ENT, heart and neurosurgery.

BIOMETRIC IDENTITY VERIFICATION - SIGNAL PROCESSING CHALLENGES

Andrzej Drygajło Ecole Polytechnique Fédérale de Lausanne, Switzerland

Biometrics is a relatively new area of technology that uses unique and measurable biological and behavioral traits of individuals to automatically establish or verify their identity. The tutorial provides a valuable insight into the state-of-the-art in biometric identity verification technology, both under the signal processing and pattern recognition point of view. It addresses the issues of the global breakthrough by biometrics in identity verification technology that is imminent in terms of its use in identity documents and corresponding applications. With an increase in identity fraud and the emphasis on security, there is a growing and urgent need to efficiently identify humans both locally and remotely on a routine basis. The appearance of electronic identity documents such as passports, visas, national identity cards, drivers' licenses and health insurance cards, have triggered a real need for reliable, user-friendly and widely acceptable automated methods for checking the identity of an individual. Biometric systems are more secure than traditional identification systems. But they only represent a secure identification process in as much as they provide a strong link between a physical person and this person identity. The attendees will have the opportunity to:

• Gain an up to date basic knowledge of the biometrics technology based on the experience of the BioSecure Network of Excellence,

• Understand the key considerations for analysis and statistical modelling of biometric digital signals, • Learn how biometrics' contribution can enhance an identity verification system, • Understand the factors affecting the performance of a biometric application, • Receive an overview of current government and private sector applications, such as Biometric Passports and

National Identity Cards, based on the experience of the European COST 2101 Action "Biometrics for Identity Documents and Smart Cards",

• Appreciate the legal and privacy issues involved. The tutorial will cover the following topics:

• Fundamentals of Biometrics Identity and Biometrics, Individuality of Biometric Data, Recognition, Verification, Identification and Authentication

• Challenges in Analysis, Modeling and Interpretation of Biometric Signals Mathematical Tools, Sensing and Storage, Representation and Feature Extraction, Enrollment and Template Creation, Statistical Models, Biometric System Errors,

• Evaluation of Biometric Systems • Leading Biometric Technology

Biological Characteristics (fingerprints, face (2D and 3D), hand geometry and veins, palmprint, iris), Behavioral Characteristics (dynamic signature, voice, gait), Technologies under Development, Synthetic Biometric Data Generation

• Multimodal Biometrics • Robustness and Reliability in Biometrics • Integration of Biometrics with other Existing Technologies (identity documents, smartcards, mobile phones, e-

technologies, transmission of biometric data) • Biometric Applications • Privacy and Legal Issues

Andrzej Drygajło, head of the Speech Processing and Biometrics Group at the Swiss Federal Institute of Technology at Lausanne (EPFL), conducts research on technological, methodological and legal aspects of biometrics for security and forensic applications.

In 1993 he created the EPFL Speech Processing Group (GTP) and then the EPFL Speech Processing and Biometrics Group (GTPB) and Biometrics Centre Lausanne. His research interests include biometrics, speech processing and man-machine communication applications. He conducts research and teaches at the School of Engineering in EPFL and at the School of Criminal Sciences in the University of Lausanne.

He participates in and coordinates numerous national and international projects and is member of various scientific committees. Among ongoing European research projects the most relevant are the Network of Excellence "BioSecure" and COST 2101 Action "Biometrics for Identity Documents and Smart Cards". Recently, he has been elected chairman of the COST 2101 Action. The main objective of this European Action is to investigate novel technologies for unsupervised multimodal biometric authentication systems using a new generation of biometrics-enabled identity documents and smart cards.