Plenary Speakers

Martin Corless

Professor, School of Aeronautics and Astronautics, Purdue University, United States.
Adjunct Honorary Professor, Hamilton Institute, National University of Ireland, Maynooth.

Title: An Approach to the Analysis and Control of Nonlinear/Uncertain Systems

Abstract: Several issues in the analysis and control of systems which are nonlinear and/or contain uncertainty in their description are considered. These issues include stability, rejection of constant and periodic disturbances, and disturbance attenuation in the presence of unknown but bounded disturbances. We consider an approach in which the nonlinear/uncertain terms in the system description are characterized by symmetric matrices called multiplier matrices; we also show that many commonly encountered nonlinear/uncertain terms can be characterized in this fashion. By employing quadratic Lyapunov functions to guarantee desired system behavior, we reduce many analysis and control design problems to that of solving a bunch of linear matrix inequalities; such inequalities can readily be solved using commercially available software.

Martin CorlessBiography: Martin Corless is currently a Professor in the School of Aeronautics & Astronautics at Purdue University, West Lafayette, Indiana, USA. He is also an Adjunct Honorary Professor in the Hamilton Institute at The National University of Ireland, Maynooth, Ireland. He received a B.E. from University College Dublin, Ireland and a Ph.D. from the University of California at Berkeley; both degrees are in mechanical engineering. He is the recipient of a National Science Foundation Presidential Young Investigator Award. His research is concerned with obtaining tools which are useful in the robust analysis and control of systems containing significant uncertainty and in applying these results to aerospace and mechanical systems and to sensor and communication networks.

Mike Davies

Professor of Signal and Image Processing
Institute for Digital Communications (IDCOM) & Joint Research Institute for Signal and Image Processing
School of Engineering and Electronics, University of Edinburgh, Scotland

Title: Compressed sensing: exploiting sparsity and structure in signal acquisition

Abstract: A new way of thinking about sensing signals has recently emerged called compressed sensing. The basic philosophy is to try to sample close to the information rate of the signal rather than the traditional Shannon/Nyquist rate. This talk will review some of the underlying theory that explains why under certain circumstances this is possible and look at some of the practical algorithms that have been developed to realize this possibility. I will then discuss some of the engineering challenges that arise in compressed sensing giving examples from advanced medical imaging, remote sensing and sub-Nyquist A2D conversion.

Mike DaviesBiography: Prof. Mike Davies holds a SFC-funded chair in Signal and Image Processing at IDCOM and is director of the Joint Research Institute in Signal and Image Processing, part of the Edinburgh Research Partnership between the University of Edinburgh and Heriot Watt University. Prof. Davies has been working in signal processing and nonlinear modelling since 1989 and specialises in nonlinear, non-Gaussian and sparse signal processing models. He has made a number of key contributions to the fields of nonlinear time series analysis, independent component analysis and sparse representations.

In more recent years Prof. Davies has been pioneering the use of sparse representations in signal processing, in particular to the problems of blind source separation and compressed sensing, giving one of the keynote talks at the first Workshop on Signal Processing with Adaptive Sparse Structured Representations. He currently leads the Edinburgh Compressed Sensing Research Group which is run in collaboration with Mathematics. He is most noted for his work on the use of the general union of subspace model in compressed sensing and on the development and analysis of the popular iterative hard thresholding algorithm. His group is further exploring the use of compressed sensing in a range of applications from advanced MRI imaging to wideband spectral sensing. For more information on his research group activities see:

Bernard Fleury

Professor of Communication Theory
Department of Electronic Systems, Aalborg University, Denmark

Title: Message Passing in Wireless Communications - Selected Applications and Recent Progress

Abstract: Belief propagation and mean-field approximation, two well-known methods of variational Bayesian inference, have found a wide application in many fields of natural and engineering sciences. They are extensively used in wireless communications, where they are commonly referred to as turbo-processing, iterative information processing, or message passing. The prefix "turbo" emphasises both the iterative (i.e. feedback) structure of these schemes and their drastic performance gain compared to traditional solutions. The term "message passing" stresses the fact that messages are exchanged between the nodes of the factor-graph describing the underlying probabilistic model. The computational power of today's DSPs and FPGAs has made the implementation of turbo-algorithms in wireless systems a reality. Classical examples are: turbo-decoding (of turbo-codes), turbo-equalization, and turbo-synchronization.

Today, the creativity of design engineers seems to be the main limiting factor to the multiplication of the applications of message-passing in wireless communications: advanced receivers embedding their key functionalities (i.e., channel estimation, channel equalization, interference cancellation, and decoding) in an iterative structure, sparse channel estimation, distributed signal processing in cooperative networks (e.g. for localization), just to mention a few of them. In this talk we will present selected examples of such recently investigated applications.

The talk will also address some recent theoretical results on message passing. We will show how the region-graph concept leads to the definition of a unified framework in which belief propagation and mean field approximation can be combined. We will also discuss the monotonicity property of variational expectation-maximization, which is an instance of mean field approximation. These results and their implications will be illustrated by means of the presented applications.

Bernard FleuryBiography: Bernard H. Fleury (M'97-SM'99) received the diploma in electrical engineering and mathematics in 1978 and 1990, respectively, and the Ph.D. degree in electrical engineering from the Swiss Federal Institute of Technology Zurich (ETHZ), Switzerland, in 1990.

Since 1997, he has been with the Department of Electronic Systems, Aalborg University, Denmark, as a Professor of communication theory. He is the Head of the Section Navigation and Communications, which is one of the eight laboratories of this department. From 2006 to 2009, he was a Key Researcher with Telecommunications Research Center Vienna (FTW), Austria. During 1978-1985 and 1992-1996, he was a Teaching Assistant and a Senior Research Associate, respectively, with the Communication Technology Laboratory, ETHZ. Between 1988 and 1992, he was a Research Assistant with the Statistical Seminar at ETHZ.

Prof. Fleury's research interests cover numerous aspects within communication theory and signal processing, mainly for wireless communications. His current research interests include stochastic modeling and estimation of the radio channel especially for MIMO applications in fast time-varying environments, iterative message-passing processing with focus on the design of efficient feasible architectures for wireless receivers, localization techniques in wireless terrestrial systems, and radar signal processing. He has authored and coauthored more than 110 publications in these areas. He has developed with his staff a high-resolution method for the estimation of radio channel parameters that has found a wide application and has inspired similar estimation techniques both in academia and in industry.

Rahul Sarpeshkar

Associate Professor of Electrical Engineering
Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, United States

Title: Ultra Low Power Biomedical and Bio-inspired Systems.

Abstract: Nature is a great analog and digital circuit designer. She has innovated circuits in the biochemical, biomechanical, and bioelectronic domains that operate very robustly with highly imprecise parts and with incredibly low levels of power. This talk will discuss how analog, RF, and bio-inspired circuits and architectures have led to and are leading to novel systems for ultra-low-power biomedical applications. Examples from systems for bionic ear processors for the deaf, brain-machine interfaces for the blind and paralyzed, body sensor networks for cardiac monitoring, and in circuits for systems biology and synthetic biology will be presented.

Rahul SarpeshkarBiography: Rahul Sarpeshkar obtained Bachelor's degrees in Electrical Engineering and Physics at MIT. After completing his PhD at CalTech, he joined Bell Labs as a member of the technical staff. Since 1999, he has been on the faculty of MIT's Electrical Engineering and Computer Science Department, where he heads a research group on Analog VLSI and Biological Systems ( He holds over 25 patents and has authored more than 100 publications, including one that was featured on the cover of Nature. His book, Ultra Low Power Bioelectronics: Fundamentals, Biomedical Applications, and Bio-inspired Systems was released in February 2010 and contains a broad and deep treatment of the fields of bioelectronics and ultra low power electronics. He has won several awards for his interdisciplinary bioengineering research including the Packard Fellow award given to outstanding faculty.

The Institution of Engineering and Technology Analog Devices MIDAS Ireland IMEX IEEE Computational Intelligence Society
Science Gallery Cypress Semiconductor Movidius Xilinx Science Foundation Ireland