Keynote Speakers

 

Christine GUILLEMOT, INRIA Rennes, France

Title: Data dimensionality reduction and neighbour embedding: algorithms and image processing applications

Images or image texture patches are high-dimensional data which usually lie on lower-dimensional subspaces, or on a non-linear manifold embedded within the high dimensional space. Finding low-dimensional structures in high dimensional data is thus at the core of the data dimensionality reduction problem which consists in searching for a mapping from the high-dimensional space to a lower-dimensional space which would preserve various input data characteristics (variance, pairwise distances, neighborhood, geometry, ..). Classical techniques for data dimensionality reduction can be classified into linear methods, e.g. PCA, which project the data into a lower dimensional space via a linear transformation, or in methods based on low rank approximations of the input data, e.g. the non-negative matrix factorization (NMF) framework. Another family of methods is based on the notion of manifold and on manifold learning which also aims at finding the low-dimensional structures in the high-dimensional data. This family of non linear methods aims at preserving local distances: E.g., multidimensional scaling (MDS) projects points while preserving the pair-wise distances between input points, locally linear embedding (LLE) aims at preserving local geometric properties. In LLE, each data point on the manifold is approximated with a linear combination of its neighbors, and the same weights are then used to compute the low-dimensional embedding, hence the name of neighbor embedding for the first step which computes the linear approximation weights.

This keynote will give an overview of algorithms for data dimensionality reduction and will in particular focus on non-linear methods. The neighbor embedding step actually turns out to be a least squares problem which can be solved under various constraints (sparsity, that the weights sum to 1, or are non-negative). Different constrained least squares formulations will be considered and corresponding algorithms (LLE, NMF with multiplicative updates, NMF with projected gradients, Graph-regularized NMF, …) will be presented. The talk will then present applications of the above algorithms for image processing problems (e.g. prediction, inpainting, super-resolution).

 

Christine Guillemot is currently ‘Directeur de Recherche’ at INRIA, in charge of a research team (TEMICS) working in the area of image and video modelling, processing, coding and communication. She holds a PhD degree from ENST (École Nationale Supérieure des Télécommunications) Paris as well as a “Habilitation à Diriger la Recherche” from the University of Rennes 1. From 1985 to October 1997, she has been with FRANCE TELECOM, where she has been involved in projects in the domain of image and video coding for TV and HDTV. She has set-up and managed EU and national projects in the area of multimedia and video processing. From January 1990 to mid 1991, she has worked at Bellcore (Bell Communication Research, NJ, USA, as a visiting scientist. She has (co)-authored 15 patents, 8 book chapters, 50 journal papers and 140 conference papers. She has served as associated editor for the IEEE Transactions on Image processing (2000-2003), for IEEE Transactions on Circuits and Systems for Video Technology (2004-2006) and for IEEE Transactions on Signal Processing (2007-2009). She is currently associate editor for the Eurasip journal on image communication. She has been a member of the IEEE IMDSP and MMSP technical committees and serves on the committee assigning the best paper award of the EURASIP Journal on Image Communication (JoIC).

Mérouane DEBBAH, Alcatel-Lucent Chair on Flexible Radio, Supélec, France

Title: Future Dense Networks: more base stations or more antennas?

Wireless networks are inherently limited by their own interference. Therefore, a lot of research focuses on interference reduction techniques, such as mutiuser MIMO, interference alignment, interference coordination or multi-cell processing. Although these techniques might lead to considerable performance gains, it is unlikely that they will be able to meet the demand for wireless data traffic in the future. Therefore, a significant network densification, i.e., increasing the number of antennas per unit area, is inevitable. One way of densifying the network consists in cell-size shrinking, such as the deployment of femto or small cells, which comes at the cost of additional equipments and increased interference. Another much simpler, but also less explored, option is the use of massively more antennas at each base station (BS). In this talk, we will discuss the challenges of small cell versus massive MIMO networks.

Mérouane Debbah entered the Ecole Normale Supérieure de Cachan (France) in 1996 where he received his M.Sc and Ph.D. degrees respectively. He worked for Motorola Labs (Saclay, France) from 1999-2002 and the Vienna Research Center for Telecommunications (Vienna, Austria) from 2002-2003. He then joined the Mobile Communications department of the Institut Eurecom (Sophia Antipolis, France) as an Assistant Professor. Since 2007, he is a Full Professor at Supelec (Gif-sur-Yvette, France), holder of the Alcatel-Lucent Chair on Flexible Radio. His research interests are in information theory, signal processing and wireless communications. He is an Associate Editor for IEEE Transactions on Signal Processing. Mérouane Debbah is the recipient of the "Mario Boella" prize award in 2005, the 2007 General Symposium IEEE GLOBECOM best paper award, the Wi-Opt 2009 best paper award, the 2010 Newcom++ best paper award as well as the Valuetools 2007, Valuetools 2008 and CrownCom2009 best student paper awards. He is a WWRF fellow. In 2011, he received the IEEE/SEE Glavieux Prize Award.

 

Mohammad GHAVAMI, LSB University, London, UK

Title: Advanced Topics in Ultra Wideband technology

Ultra wideband (UWB) radio is an emerging technology, which has thepotential to be a viable candidate to offer a unique communicationplatform for several short range applications. It operates using ultra short duration pulses that yield ultra wide bandwidth signals. In particular, UWB systems have low power and low cost and are resistant to severe multipath fading. They also provide greater bandwidth and are more cost effective for providing increased throughput without increasing power. Currently, UWB systems are the hot topic of research and development in the Europe, USA and Japan. Many companies have been involved in this technology and invested a large amount of money, which shows a great deal of hope for the near future of this technique.

This tutorial starts with an introduction which comprise basic properties of UWB signals and systems, generation of UWB waveforms, UWB channel modelling, and ultra wideband communications. Then, UWB antenna and arrays, UWB applications including medical and consumer electronics applications, and advanced recent topics in ultra wideband technology will be discussed.

 

Tayeb DENIDNI, INRS, Université de Laval, CA

Title: Modern antennas for wireless communication systems 

This tutorial will cover the fundamentals and the current and future trends related to antennas and their applications in wireless communication systems. The technical material will be application oriented and emphasize on practical aspects with supporting theory provided. The tutorial will begin with addressing the basic concepts of antennas introducing the essential and important parameters such as the radiation pattern, gain and input impedance. Examples of popular antennas (dipole, monopole, and loop antennas) will also be provided. As applications for wireless communication systems, some modern and important developments in antennas (multi-band, ultra-wideband, and directive) for base stations, vehicular and handheld terminals will be addressed. Furthermore, antenna arrays will be discussed. Special considerations for popular antenna systems including diversity antenna systems, beamforming antenna arrays, and smart antennas will be presented. Current and future trends in antenna design for wireless communication systems will be discussed at the conclusion of this tutorial.