Disaggregated sustainability sensing measures the amount of usage from each fixture or appliance in a space—for example, providing information on the amount of water, electricity, or gas use at each point in the home. “The kitchen faucet was used 33 times today, for a total of 10 gallons.” This type of information enables a number of end-use studies on sustainable water, gas, and electricity use, as well as enabling real-time eco-feedback monitoring. In this talk I will explain how we sense disaggregated usage using Infrastructure Mediated Sensing (IMS). IMS infers activity in a space or residence by sensing at a single point along a home’s infrastructure (i.e., a pressure sensor for sensing whole home water use, a single microphone for sensing gas use, a plug-in sensor for electricity). I will give an overview of each system and give an extended explanation of our work in water sensing.
Eric Larson is a senior PhD student at the University of Washington, with expected completion in April 2013. He received his M.Sc. in EE from Oklahoma State University in 2008. He is advised by Shwetak Patel in the Laboratory of Ubiquitous Computing at the UW. His dissertation focuses on signal processing and machine learning that support sustainable water use, and he is working on semi-supervised modeling to make the designs practical. The tools that he is currently developing could be used by many others who are non-experts in machine learning. He has a broad range of interests, including research in image processing, optimization, technical HCI, and, most recently, sensing markers of health from mobile phones.
Live broadcast at
. Ask questions live on Twitter: #CITRISi4E. All talks may be viewed on our
The schedule for the semester can be found on the
Webviewing at UC Davis: 1003 Kemper Hall
Webviewing at UC Merced: SE1 100
Webviewing at UC Santa Cruz: SOE E2 Building, Room 595B