EURASIP Seminar on Robust Statistics for Signal Processing

This Seminar sponsored by EURASIP is held in the framework of the First IEEE SPS Italy Chapter Summer School on Signal Processing, Riotorto (LI) - Italy - September 1-6, 2013 , organized by Prof. M. Barni (University of Siena), and Dr. A. Piva (University of Firenze).
Speaker: Prof. Abdelhak Zoubir, Technische Universität Darmstadt, Germany
Venue: Borgo degli Olivi,  Via Rosanna Benzi 12, 57025 Riotorto Livorno.
When: September 2,2013 at 9.00-12.30 
Slides: you can dowload the slides of the presentation here
Statistical signal processing often relies on strong and precise assumptions, i.e. optimal estimators, detectors and filters are derived based e.g. on a particular parametric model, a probability distribution of the sensor noise (typically the normal distribution) or the assumptions of linearity or independence and identical distribution of random variables. Optimality however, is only achieved when the underlying assumptions hold, and the performance of optimal procedures deteriorates significantly, even for minor departures from the assumed model. In particular, measurement campaigns have revealed the presence of impulsive noise, which can cause conventional signal processing techniques, especially the ones derived using the nominal Gaussian probability model, to be biased or even to break down. The occurrence of impulsive noise has been reported, for example, in outdoor mobile communication channels, due to switching transients in powerlines or automobile ignition, in radar and sonar systems as a result of natural or man-made electromagnetic and acoustic interference, and in indoor wireless communication channels, owing e.g. to microwave ovens and devices with electromechanical switches, such as electric motors in elevators, printers, copying machines etc. Moreover, biomedical sensor array measurements of the brain activity (MRI) were found to have non-Gaussian noise and interference in various regions of the human brain, where a complex tissue structure is known to exist. In geolocation position estimation and tracking, non-line-of-sight (NLOS) signal propagation, caused by obstacles such as buildings or trees, results in outliers in the measurements, to which conventional position estimation methods are very sensitive. As a final example, in classical short-term load forecasting, the prediction accuracy is adversely influenced by outliers, which are caused by non-working days or exceptional events such as strikes, the football world cup or natural disasters. Consequently, in these situations, there is a need for robust statistical signal processing methods, which are close-to optimal in nominal conditions and highly reliable for real life data, even if the assumptions are only approximately valid.
In this course, we give an overview of robust estimation and its application to signal processing. Real life examples from diverse applications, such as array processing, image restoration, geolocation and load forecasting, interwound with theoretical concepts, will be used throughout the course to illustrate the applicability and performance of various robust methods. We will address single and multi-channel estimation problems, as well as linear univariate regression for independently and identically distributed (i.i.d.) data. The challenging case of correlated data will also be treated. A comparative performance analysis of the most important robust methods will be carried out theoretically, via simulations, as well as real world experiments. Throughout the course, the attendee will be referred to available implementations of the algorithms.
Abdelhak M. Zoubir is a Fellow of the IEEE and IEEE Distinguished Lecturer (Class 2010- 2011). He received his Dr.-Ing. from Ruhr-Universität Bochum, Germany in 1992. He was with Queensland University of Technology, Australia from 1992-1998 where he was Associate Professor. In 1999, he joined Curtin University of Technology, Australia as a Professor of Telecommunications and was Interim Head of the School of Electrical & Computer Engineering from 2001 until 2003. In 2003, he moved to Technische Universität Darmstadt, Germany as Professor of Signal Processing and Head of the Signal Processing Group. His research interest lies in statistical methods for signal processing with emphasis on bootstrap techniques, robust detection and estimation and array processing applied to telecommunications, radar, sonar, automotive monitoring and safety, and biomedicine. He published over 300 journal and conference papers on these areas. Professor Zoubir was the Technical Chair of the 11th IEEE Workshop on Statistical Signal Processing (SSP 2001), General Co-Chair of the 3rd IEEE International Symposium on Signal Processing & Information Technology (ISSPIT 2003) and of the 5th IEEE Workshop on Sensor Array and Multi-channel Signal Processing (SAM 2008). He is the General Co-Chair of the 14th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC 2013) to be held in Darmstadt, Germany, the General Co-Chair of the 21st European Signal Processing Conference (EUSIPCO 2013) to be held in Marrakech, Morocco, and the Technical Co-Chair of ICASSP-14 to be held in Florence, Italy. Dr. Zoubir was an Associate Editor of the IEEE Transactions on Signal Processing (1999-2005), a Member of the Senior Editorial Board of the IEEE Journal on Selected Topics in Signal Processing (2009-2011) and he currently serves on the Editorial Boards of the EURASIP journals Signal Processing and the Journal on Advances in Signal Processing (JASP). He currently serves a three-year term as Editor-In-Chief of the IEEE Signal Processing Magazine (2012-2014). Dr. Zoubir was Past Chair (2012), Chair (2010-2011), Vice-Chair (2008-2009) and Member (2002-2007) of the IEEE SPS Technical Committee Signal Processing Theory and Methods (SPTM). He was a Member of the IEEE SPS Technical Committee Sensor Array and Multi-channel Signal Processing (SAM) from 2007 until 2012. He also serves on the Board of Directors of the European Association of Signal Processing (EURASIP).



Gruppo di ricerca: