Fredrik Bajers Vej 5
P.O. Box 159 DK-9100 Aalborg
Phone: +45 9940 9940
21.06.2017 kl. 13.00 - 14.30
“Random projections and information coding”
By Petros Boufounos, Senior Principal Research Scientist, Mitsubishi Electric Research Laboratories, USA
The emergence of compressed sensing has invigorated attention in the information content of signals, and on how this information is preserved in linear measurements. In particular, random projections preserve the information content of a signal, under a variety of signal models, as long as the number of projections obtained is sufficient for the complexity of the signal model. Furthermore, it can be shown that random projections preserve other signal space properties, such as functions of signal distances and signal correlations.
In that sense, random projections provide a very useful, low-complexity, universal encoding mechanism that can be used to compress relevant signal information. Unfortunately, in a number of cases, including standard compressive sensing problems, random projections are still very redundant and underperform compared to conventional higher complexity encoding approaches, such as transform coding.
In this talk I will demonstrate how appropriate quantization methods, often combined with distributed coding, can overcome this redundancy and provide extremely simple encoding schemes that perform as well as state-of-the-art transform coding approaches.
Petros T. Boufounos is Senior Principal Member of Research Staff and the Computational Sensing Team Leader at Mitsubishi Electric Research Laboratories (MERL), and a visiting scholar at the Rice University Electrical and Computer Engineering department. Dr. Boufounos completed his undergraduate and graduate studies at MIT. He received the S.B. degree in Economics in 2000, the S.B. and M.Eng. degrees in Electrical Engineering and Computer Science (EECS) in 2002, and the Sc.D. degree in EECS in 2006. Between September 2006 and December 2008, he was a postdoctoral associate with the Digital Signal Processing Group at Rice University. Dr. Boufounos joined MERL in January 2009.
Dr. Boufounos' immediate research focus includes signal acquisition and processing, frame theory, quantization and data representations. He is also interested into how signal acquisition interacts with other fields that use sensing extensively, such as machine learning, robotics and mechatronics. Dr. Boufounos is a Senior Area Editor at IEEE Signal Processing Letters. He is also a senior member of the IEEE and a member of Sigma Xi, Eta Kappa Nu, and Phi Beta Kappa.
All are welcome!
Free of charge
Signal Information and Processing, Department of Electronic Systems
Aalborg University, Fredrik Bajers Vej 7, B3-104