Wireless Occupancy Detection

Wireless Occupancy Detection to Improve Building Energy Efficiency

Funding: NSF (CBET 1336824

PI: Tat S. Fu; Co-PI: Nicholas Kirsch (UNH Electrical  Engineering)

Duration: 2014-2017


Buildings are the largest consumers of energy in the United States and, therefore, creating green buildings can significantly aid the country’s pressing need to enhance energy efficiency and sustainability. The objective of this project is to detect and estimate the number of occupants in buildings to control building energy-consuming systems and thereby increase energy efficiency. Conventional occupancy sensors detect the presence of occupants, where a single person is treated equally as a group of people. For building controls such as lighting, these presence-based methods are sufficient. However, controls, such as ventilation, heating and cooling, may fluctuate considerably depending on the number of occupants in a room. Instead of detecting the physical presence of occupants from movements or body heat as in conventional sensors, the investigators propose to detect occupancy by sensing existing cellphone communications. A recently published study reported that 99% of individuals carry cellphones and even idle cellphones communicate routinely with cell towers. By sensing cellphone signals inside buildings, the investigators propose that the number and locations of cellphones (and their human carriers) can be accurately determined. The occupancy data will then be provided to building automation systems to improve energy efficiency by adjusting different building systems (ventilation, heating/cooling, etc.) accordingly. This research includes three objectives: cellular communication-based occupant tracking, analysis and prediction of stochastic occupancy for energy efficient control strategies, and prototype testing in real buildings. In the first objective, the investigators will use multiple cellular phone control-channel traffic sensors to identify the position of each occupant. The second objective of the project focuses on data fusion and analysis of the cellular signal data. The investigators will develop algorithms that determine the number of occupants and position based upon the spectrum-sensing network. With occupancy tracking data, the investigators will develop stochastic models of occupancy in test buildings. Such models will be used to predicate occupancy levels for the following phase of this project. In the last objective, the investigators aim to implement new control strategies that leverage the comprehensive occupancy data to improve efficiency in building control systems. Experiments will be conducted in ten campus buildings to implement the proposed methods in collaboration with university building management.

The success of the proposed system will improve building sustainability and potentially change the fields of building design and operations by providing real-time occupancy data. Moreover, the proposed non-intrusive occupancy detection system may find use in areas such as security, rescue efforts (locating victims) and elderly care (resident tracking in nursing homes). The PIs will incorporate different aspects of the research into Electrical/Computer and Civil Engineering courses (Wireless Communication System and Green Building Design) as well as the Dual Major in Sustainability at UNH. Students from different backgrounds will work as a team and be exposed to diverse, interdisciplinary research topics.


Fu, T., Adams, T., and Kirsch, N., 2015. “Wireless Occupancy Detection with Dynamic Schedules.” Proceedings of the 2015 Conference of the ASCE Architectural Engineering Institute (AEI), Milwaukee, WI, March 24-27, 2015 (accepted).

Bera, R., Kirsch, N.J., and Fu, T., “An Indoor Probabilistic Localization Method Using Prior Information,” Proceedings of the 2013 IEEE Vehicular Technologies Conference, Dresden, Germany, June 2-5, 2013.

Bera, R., Kirsch, N.J., and Fu, T., “Using Prior Measurements to Improve Probabilistic-based Indoor Localization Methods,” Proceedings of the 2013 IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, Berlin, Germany, September 12-14, 2013.