Electronics for Interacting with a Physically-Complex World

https://www.eecs.umich.edu/eecs/etc/calendar/showcalendar.cgi?731
Think about some of the physical systems with which we would like electronics to engage in valuable interactions: physiological systems, high-value industrial equipment, critical infrastructure…. These systems are complex, both in terms of the number of signals they present, and in terms of how those signals represent information. In this talk I will describe some of the hardware platforms we are pursuing to handle these complexities. By 'interacting with a physically-complex world', I am referring to an ability to make sense of embedded signals when in fact we may have no tractable analytical models for the underlying processes. Instead, we look at how sensor data can be used as a knowledge base, exploiting the tremendous data-acquisition capabilities of next-generation sensor networks towards the construction of high-quality data-driven models. Machine learning gives us powerful frameworks for data-driven analysis; the question is how to create very-low-power hardware to enable such frameworks in energy-constrained sensor devices. I will describe our work on low-power medical sensors for disease monitoring and harm detection. Interacting with a physically-complex world also implies the ability to acquire embedded signals on a very large scale, in fact much larger than traditional integrated-circuit technologies can possibly handle. Large-area electronics is a technology that can enable the creation of large arrays of diverse transducers for sensing. To build complete systems, however, substantial embedded computation, instrumentation, and power management is also required. We focus on scalable methods for combining large-area electronics with CMOS ICs to exploit the complementary strengths of the two technologies. I will describe our work towards smart infrastructure, using flexible sensing sheets to build systems for high-resolution structural-health monitoring of bridges.
Naveen Verma received the B.A.Sc. degree in Electrical and Computer Engineering from the University of British Columbia, Vancouver, Canada in 2003 and the M.S. and Ph.D. degrees in Electrical Engineering from Massachusetts Institute of Technology in 2005 and 2009 respectively. Since July 2009 he has been an Assistant Professor of Electrical Engineering at Princeton University. His research focuses on advanced sensing systems, including low-voltage digital logic and SRAMs, low-noise analog instrumentation and data-conversion, large-area sensing arrays based on flexible electronics, and low-energy algorithms for embedded inference especially for medical applications.