Sentinel: Occupancy Based HVAC Actuation using Existing WiFi Infrastructure within Commercial Buildings

Commercial buildings contribute to 19 % of the primary energy consumption in the US, with HVAC systems accounting for 39.6% of this usage. To reduce HVAC energy use, prior studies have pro-posed using wireless occupancy sensors or even cameras for oc-cupancy based actuation showing energy savings of up to 42%. However, most of these solutions require these sensors and the as-sociated network to be designed, deployed, tested and maintained within existing buildings which is significantly costly. We present Sentinel, a system that leverages existing WiFi in-frastructure in commercial buildings along with smartphones with WiFi connectivity carried by building occupants to provide fine-grained occupancy based HVAC actuation. We have implemented Sentinel on top of RESTful web services, and demonstrate that it is scalable and compatible with legacy building management. We show that Sentinel accurately determines the occupancy in office spaces 86 % of the time, with 6.2 % false negative errors. We high-light the reasons for the inaccuracies, mostly attributed to aggres-sive power management by smartphones. Finally, we actuate 23% of the HVAC zones within a commercial building using Sentinel for one day and measure HVAC electrical energy savings of 17.8%. Categories and Subject Descriptors [Computer systems organization]: Special purpose systems, sen-sors and actuators, real-time system architectur