People/Web Search Calendar Emergency Info A-Z Index UVA Email University of Virginia

Computer Science Colloquia

Friday, January 17, 2014

Prabal Dutta, University of Michigan

3:30 PM, Rice Hall, Rm. 130

HOST: Jack Stankovic

Scalable Sensor Infrastructure for Sustainably Managing the Built Environment


Federal sustainability goals mandate that 50% of U.S. commercial buildings become net-zero energy by 2050. A range of options exists to achieve this goal, but financial concerns require a data driven and empirically validated approach. However, critical gaps exist in the measurement technology and indoor climate control science needed to benchmark competing options, prioritize efficiency investments, and ensure occupant comfort. To help address these challenges, we are exploring new techniques for resource monitoring, including new sensors that can be attached to everyday objects, like ceiling lights, shower heads, circuit panels, or range tops, to infer their contributions to whole-building resource consumption, or placed in the vicinity of loads to be monitored. The sensors monitor the ambient conditions around a load and, using statistical methods, correlate those conditions with readings from existing electricity, gas, or water meters, providing individual estimates, often without intrusive metering. The insight underlying this work is that the transfer and use of energy (and other resources) usually emits energy, often in a different domain, and that this emitted energy is often enough to intermittently power simple, energy harvesting sensors whose duty cycle is proportional to the energy being transferred or used. We demonstrate the viability of the concept, discuss some potentials drawbacks and opportunities for improvement, and present our roadmap for scaling the sensors.

Bio: Prabal Dutta is an Assistant Professor of Electrical Engineering and Computer Science at the University of Michigan, Ann Arbor. He researches the circuits, systems, and software necessary to realize pervasive sensing, computing, and communications at scale and in the service of society. His work has yielded over a dozen hardware and software systems, has won four best paper awards (Sensys'10, IPSN'10, HotEmNets'10, and IPSN'08) and several design awards (ISLPED'10, ISLPED'08, and Comdex Best of Show), has been directly commercialized by a dozen companies and indirectly by dozens more, and has been utilized by thousands of researchers and practitioners worldwide. His work has been recognized with an NSF CAREER award and an Intel Early Career Award. He earned a Ph.D. in Computer Science from the University of California, Berkeley, where he designed and deployed some of the largest academic sensor networks, and where his graduate research was supported by NSF and Microsoft Graduate Fellowships. He also holds a B.S. in Electrical and Computer Engineering and an M.S. in Electrical Engineering from The Ohio State University.