1;2c1;2c Grants :: NSF CNS-1350967
Enabling Smarter and More Efficient Buildings

Scalable Sensor Infrastructure for Sustainably Managing the Built Environment

Duration: February 1, 2014 → January 31, 2019

Synopsis U.S. economic growth, energy security, and environmental stewardship depend on a sustainable energy policy that promotes conservation, efficiency, and electrification across all major sectors. Buildings are the largest sector and therefore an attractive target of these efforts: current 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, empirically-validated approach. However, critical gaps exist in the energy and water measurement technology, and indoor climate control science, needed to benchmark competing options, prioritize efficiency investments, and ensure occupant comfort.

To address these challenges, this project proposes a new kind of "peel-and-stick" sensor that can be affixed to everyday objects to infer their contributions to whole-building resource consumption. To use the sensors, occupants or building managers simply tag end loads like a ceiling light, shower head, or range top. 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 without intrusive metering. The sensors are built from integrated circuit technology laminated into smart labels, so they are small, inexpensive, and easy-to-deploy. The sensors are powered by the same ambient signals they sense, eliminating the need for periodic battery replacement or wall power. Collectively, these properties address cost and coverage challenges, and enable scalable deployment and widespread adoption. Coupled with intelligent lighting, heating, ventilation, and air conditioning, these enabling technologies will lead to smarter, more efficient, and more responsive environments.

The intellectual merit of this project stems from the key insight 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. Hence, the mere activation rate of the sensors signals the underlying energy use. The power-proportional relationship between usage activity and side channel harvesting, when coupled with state-of-the art, millimeter-scale, nano-power chips and whole-house or panel-level meters, enables small and inexpensive sensor tags that are pervasively distributed with unbounded lifetimes. But, networking and tasking them, and making sense of their data, requires a fundamental rethinking of low-power communications, control, and data fusion to abstract the intermittent, unreliable, and noisy sensor infrastructure into actionable information.

This project's broader impacts stem from an integrated program of education, research, and outreach that (i) creates a smart objects focused curriculum whose classroom projects are motivated by research needs, (ii) provides research experiences for undergraduates and underrepresented minorities, (iii) mentors students on all aspects of successful research from articulating hypotheses to peer-reviewing papers, (iv) disseminates teaching materials on embedded systems and research pedagogy, (v) produces students who bridge disciplines, operating at the intersection of measurement science, information technology, and sustainability policy, (vi) translates scientific discovery and technical knowledge into beneficial commercial products through industry outreach and internships, and (vii) engages with the National Labs to ensure that the research addresses pressing problems.

Elements of the Modular Monjolo Measurement System.


Disaggregating Power with Monjolo and Deltaflow.

Demos Some real-time data streams collected using the Monjolo sensors:

People Members of the core project team from Michigan's Lab 11 include:

Collaborators We're working with the following third parties on this work:

Education The key educational activity supported by the project is the introduction of this research into the EECS 373 course "Design of Microprocessor-Based Systems" at the University of Michigan. This has enabled multiple student teams to design hardware and software that explore the key ideas underlying this proposal, including embedded systems, energy harvesting, low-power sensing, and data processing. Some students engage in undergraduate research experiences after the class project to continue researching and developing their class projects, characterizing the project's performance, publishing the results, presenting academic papers at research conferences, and (often) continuing to graduate school (sometime with competitve fellowships including NSF and NDSEG).

Our HW/SW designs are available from the Lab 11 website and from GitHub.

  • CoilCube: A Monjolo magnetic energy harvester layer.
  • Gecko: A Monjolo solar energy harvester layer.
  • HotSpring: A Monjolo thermoelectric energy harvester layer.

  • Impulse: A Monjolo MSP430/802.15.4-based MCU/radio/memory layer.
  • sEHnsor: A Monjolo ARM-CM3/802.15.4-based MCU/radio/memory layer.
  • Blast: A Monjolo nRF51822/BLE-based MCU/radio/memory layer.

  • BigBen: Time-keeping and FeRAM NV storage for Impulse board stack.
  • Breeze: An indoor airflow sensor.

  • Bump: A vibration motion trigger for Monnjolo.
  • Buzz: A piezo-electric motion trigger for Monnjolo.

  • CoilCube: An integrated Monjolo power meter.

Source code for artifacts produced under this project include:

  • Monjolo: A modular, energy-harvesting, energy metering platform.
  • Harmonia: A system for localizing tiny tags indoors.
  • Luxapose: A system for localizing phones indoors.
  • Opo: A system for localizing people and tracking interactions indoors.
Presentations Past (and planned future) non-conference presentations related to this project:

  1. Prabal Dutta, "Realizing the Vision of Perpetual Smart Dust," CMOS Emerging Technologies Research, Vancouver, BC, Canada, May 20-22, 2015. Upcoming.

  2. Prabal Dutta, "Scalable Sensor Infrastructure for Sustainably Managing the Built Environment," Distinguished Lecture Series in Computer Science, University of Virginia, Charlottesville, VA, Dec. 11, 2014. Upcoming.

  3. Prabal Dutta, "Keynote Address," ENSsys’14: The 2nd International Workshop on Energy Neutral Sensing Systems, Memphis, TN, Nov. 6, 2014.

  4. Prabal Dutta, "Monitoring Electricity and the Environment with Mobile Phones," UC Berkeley/World Bank CEGA-DIME Workshop on Measuring Development: Energy & Environment, Berkeley, CA, Aug. 18, 2014.

  5. Pat Pannuto, "Embedded System Design and the Internet of Things," Secure Internet of Things Project's 2014 Industrial Workshop, Stanford, CA, Aug. 11, 2014.

  6. Prabal Dutta, "Realizing the Next Computing Class: From Pervasive to Perpetual Computing," Ultra-Low Power Computing Workshop, Microsoft Research, Redmond, WA, Jul. 16, 2014.

  7. Prabal Dutta, "Smart Dust Sensor Networks," PLM14101: Global Information Technology Outlook, World Economic Forum’s Forum Academy, Jun. 17 – Jul. 29, 2014. (approximately 60 participants).

  8. Meghan Clark, "Deltaflow: Submetering by Synthesizing Uncalibrated Pulse Sensor Streams," University of Michigan CSE Prelim Exam, May 2014.

  9. Prabal Dutta, "Realizing the Next Computing Class," Workshop on Connected, Autonomously Powered Systems at Columbia University, New York, NY, Apr. 11, 2014.

  10. Prabal Dutta, "Towards Smarter and More Efficient Buildings," Texas Instruments, Dallas, TX, Mar. 21, 2014.

  11. Prabal Dutta and Anthony Rowe, "Localization Services and Semantic Localization," Intel Corporation, Portland, OR, Mar. 14, 2014.

  12. Pat Pannuto, "Sensing Technologies for Data Collection and Monitoring", The 1st Annual State of the Science Conference, Jointly hosted by DIL, gui2de, CEGA, and USAID, Washington D.C., Mar. 7, 2014.
Publications This project has supported the following publications:

⟨⟨ 2014 ⟩⟩

  1. "Gemini: A Non-Invasive, Energy-Harvesting True Power Meter,"
    Bradford Campbell and Prabal Dutta,
    In Proceedings of the 35th IEEE Real-Time Systems Symposium (RTSS'14),
    Rome, Italy, Dec. 2-5, 2014. To appear.

  2. "Energy-Harvesting Thermoelectric Sensing for Unobtrusive Water and Appliance Metering,"
    Bradford Campbell, Branden Ghena, and Prabal Dutta,
    In Proceedings of the 2nd International Workshop on Energy Neutral Sensing Systems (ENSsys'14),
    Memphis, TN, Nov. 6, 2014. To appear.

  3. "An Energy-Harvesting Sensor Architecture and Toolkit for Building Monitoring,"
    Bradford Campbell and Prabal Dutta,
    In Proceedings of the 1st ACM International Conference on Embedded Systems For Energy-Efficient Buildings (BuildSys'14),
    Memphis, TN, Nov. 5-6, 2014. To appear.

  4. "Opo: A Wearable Sensor for Capturing High-Fidelity Face-to-Face Interactions,"
    William Huang, Ye-Sheng Kuo, Pat Pannuto, and Prabal Dutta,
    In Proceedings of the 12th ACM Conference on Embedded Networked Sensor Systems (Sensys'14),
    Memphis, TN, Nov. 3-6, 2014.

  5. "MBus: A 17.5 pJ/bit/chip Portable Interconnect Bus for Millimeter-Scale Sensor Systems with 8 nW Standby Power,"
    Ye-sheng Kuo, Pat Pannuto, Gyouho Kim, Zhiyoong Foo, Inhee Lee, Ben Kempke, Prabal Dutta, David Blaauw, and Yoonmyung Lee,
    In IEEE Custom Integrated Circuits Conference (CICC'14),
    San Jose, CA, Sep. 14-17, 2014.

  6. "Harmonia: Wideband Spreading for Accurate Indoor RF Localization,"
    Benjamin Kempke, Pat Pannuto, and Prabal Dutta,
    In The 1st ACM Workshop on Hot Topics in Wireless (HotWireless'14),
    Maui, HI, Sep. 11, 2014.

  7. "Luxapose: Indoor Positioning with Mobile Phones and Visible Light,"
    Ye-Sheng Kuo, Pat Pannuto, Ko-Jen Hsiao, and Prabal Dutta,
    In Proceedings of the 20th Annual International Conference on Mobile Computing and Networking (MobiCom'14),
    Maui, HI, Sep. 7-11, 2014.

  8. "System Architecture Directions for a Software-Defined Lighting Infrastructure,"
    Ye-Sheng Kuo, Pat Pannuto, and Prabal Dutta,
    In The 1st ACM Workshop on Visible Light Communication Systems (VLCS'14),
    Maui, HI, Sep. 7, 2014.

  9. "Deltaflow: Submetering by Synthesizing Uncalibrated Pulse Sensor Streams,"
    Meghan Clark, Bradford Campbell, and Prabal Dutta,
    In Proceedings of the 5th ACM International Conference of Future Energy Systems (e-Energy'14),
    Cambridge, UK, Jun. 11-13, 2014.
Prior Related Work Our closely-related prior work in this areas includes:

  1. "Monjolo: An Energy-Harvesting Energy Meter Architecture,"
    Samuel DeBruin, Bradford Campbell, Prabal Dutta,
    In Proceedings of the 11th ACM Conference on Embedded Networked Sensor Systems (Sensys'13),
    Rome, Italy, Nov. 11-14, 2013.

  2. "Grafting Energy-Harvesting Leaves onto the Sensornet Tree,"
    Lohit Yerva, Bradford Campbell, Apoorva Bansal, Thomas Schmid, and Prabal Dutta,
    In Proceedings of the 11th International Conference on Information Processing in Sensor Networks (IPSN'12),
    Beijing, China, Apr. 16-20, 2012.

  3. "Exploring Powerline Networking for the Smart Building,"
    Pat Pannuto and Prabal Dutta,
    In Extending the Internet to Low power and Lossy Networks (IP+SN'11),
    Chicago, IL, United States, Apr. 11, 2011.

  4. "Meter Any Wire, Anywhere by Virtualizing the Voltage Channel,"
    Thomas Schmid, David Culler, and Prabal Dutta,
    In 2nd ACM Workshop On Embedded Sensing Systems For Energy-Efficiency In Buildings (Buildsys'10),
    Zurich, Switzerland, Nov. 2, 2010.

  5. "Disentangling Wireless Sensing from Mesh Networking,"
    Thomas Schmid, Roy Shea, Mani Srivastava, and Prabal Dutta,
    In Workshop on Hot Topics in Embedded Networked Sensors (HotEmNets'10),
    Killarney, Ireland, Jun. 28-29, 2010. Best Paper Award.
This material is based upon work supported by the National Science Foundation under award #1350967 (CNS-CPS). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

CSE Division
EECS Department
University of Michigan
2260 Hayward Street
Ann Arbor, Michigan 48109