DATE:

Friday, April 9, 2021

TIME:

3:30 PM

LOCATION:

Virtual Zoom Conference

SPEAKER:

Hakan Toreyin, Electrical and Computer Engineering, San Diego State University

ABSTRACT:

The increasing number of nodes in a sensor network imposes significant power burden on the network. This burden is primarily associated with transmission and processing of the vast amount of data that the network generates. An approach to reduce the data volume, and thus the total power consumption, is adding intelligence to the nodes. In the 1st half of the talk, energy-efficient computation electronics using physical principles to process continuous time signals, will be presented in the context of biosignal processing and body-sensor-networks.

Suffering from the von Neumann- (memory-) bottleneck, existing digital processors do not offer the most power- or speed- efficient solutions for machine learning applications. In the 2nd half of the talk, new hardware accelerator approaches will be outlined, which will be followed by a discussion on neuromorphic and analog/mixed-signal approaches.

Bio: Hakan Töreyin is an assistant professor at the department of Electrical and Computer Engineering at the San Diego State University. He received the B.S. degree in electrical and electronics engineering from the Middle East Technical University, Ankara, Turkey in 2007; the M.S. and Ph.D. degrees in electrical and computer engineering from Georgia Institute of Technology, Atlanta, GA in 2008 and 2014, respectively. Dr. Töreyin was a Fulbright Scholar during his M.S. studies. He received Chih Foundation research award in 2012. He was the North America Geographic Finalist and awarded the third price in the Student Paper Competition at the IEEE EMBC 2014. Dr. Töreyin is an associate editor of IEEE Journal of Translational Engineering in Health and Medicine and IEEE Transactions on Biomedical Circuits and Systems.

HOST:

Christopher Paolini

VIDEO: