Compressive Sensing and Beyond: New Approaches to Signal Acquisition
<-- Return to the list
Date: 02-19-2008
Start Time:
11:00am
End Time: 12:00pm
Speaker: Petros Boufounos
From:
Rice University
Location: Interschool Lab, Room 750 CEPSR
Abstract
Recent theoretical results on signal acquisition enable us to robustly acquire, represent, reconstruct, and process signals much more efficiently than previously. For example, Compressive Sensing exploits the compressibility of a wide variety of signals in linearly acquiring data at a rate significantly lower than the Nyquist rate. This approach exploits randomness, signal structure, and increasingly inexpensive computation to significantly improve signal acquisition performance.
This talk provides an overview of the theoretical advances along with specific applications where they show significant promise. We demonstrate that consistent reconstruction methods can be incorporated in the Compressive Sensing framework to acquire and reconstruct signals using only 1-bit compressive measurements. For instance in a particular imaging problem, we show that incorporating simple reconstruction constraints leads to robust acquisition at a rates as low as 1/8 bits per pixel. Similarly, sparsity assumptions
and optimization-based reconstruction allow us to acquire and robustly reconstruct signals by detecting only their zero crossings. The unifying theory behind these new applications provides significant insight into the future directions in signal acquisition and processing.
Bio
Petros Boufounos completed his undergraduate and graduate studies in the Massachusetts Institute of Technology (MIT) and 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. Since September 2007, Dr. Boufounos has been with the Rice University Digital Signal Processing Group doing research in the area of compressive sensing. In addition to compressive sensing, his research interests include signal processing, data representations, frame theory, and machine learning applied to signal processing.Dr. Boufounos has received the Ernst A. Guillemin Master Thesis Award for his work on DNA sequencing and the Harold E. Hazen Award for Teaching Excellence, both from the MIT EECS department. He has also been an MIT presidential fellow. Dr. Boufounos is a member of the IEEE, Sigma Xi, Eta Kappa Nu, and Phi Beta Kappa.