The UCL/ICTEAM IEEE Student Branch along with prof. David Bol is glad to invite you to the following seminar :
“Analog-to-information sensing, going beyond compressive sampling”
given by Marian Verhelst, SSCS Distinguished Lecturer and professor at KU Leuven (MICAS group).
The event will be held on Monday 2nd of May at 2:00 pm in the Shannon Room(Maxwell A.105).Drinks and appetizers will be served after the presentation and the questions-and-answers session.
Registration is free but mandatory, please sign up at the bottom of this page.
Abstract: the number of sensors in our devices is currently exploding. Additionally, a second, crucial trend is jointly arising: We are no longer interested in the exact signal waveforms stemming from these sensors. In many applications, not the waveform itself, but only the information hidden in this waveform is what counts. This so called “information rate” can be well below the Nyquist rate of the signal. The fundamental question that arises is how we can extract all relevant information, while spending minimal (energy) resources on irrelevant data.
Compressed sensing has recently caught much attention to pursue this goal. However, many other complementary techniques are arising, resulting in equally good, if not even better information extraction efficiency. We will review several of these techniques, and assess their utility in function of the signal characteristics of the waveform under observation, and nature of the relevant information fraction. This will be illustrated based on practical examples from the acoustic and vision sensing domain.
Biography: Marian Verhelst is a professor at the MICAS group of the Electrical Engineering Department of KU Leuven, which houses KU Leuven’s research on integrated circuit design. Her research focuses on processing architectures for ubiquitous sensing and machine learning, as well as on self-adaptive sensor interfaces for the internet-of-things. Before that, she received a PhD from KU Leuven, was a visiting scholar at the Berkeley Wireless Research Center (BWRC) of UC Berkeley and worked for 3 years in the research labs of Intel Corporation in Portland, OR, USA. Marian has a passion for interdisciplinary collaborations and for science communication. She is a member of the Young Academy of Belgium.