Computer Science Location: Link Lab 211
Add to Calendar 2022-12-14T11:30:00 2022-12-14T11:30:00 America/New_York Ph.D. Qualifying Exam Presentation by Fateme Nikseresht Adaptable Sensor-free Occupancy Detection using WiFi Fine Time Measurement Abstract:   Link Lab 211

Adaptable Sensor-free Occupancy Detection using WiFi Fine Time Measurement

Abstract:  

Occupancy sensing offers crucial information for energy management and cost reduction in buildings. The growing trend of utilizing low-cost and low-power IoT devices in smart buildings and the recent advancements in radio frequency~(RF) chips have opened up new potential for accurate sensor-free occupancy detection in indoor spaces. Among the current RF-based technologies for occupancy monitoring, WiFi-based approaches are the most promising ones because WiFi infrastructure is widely available and easily accessible in smart buildings. However, the current WiFi-based occupancy sensing systems' performance is sensitive to changes in the physical space configuration. In addition, their performance might be affected by signal strength fluctuation due to interference with other WiFi signals in the environment. To alleviate these shortcomings, I propose an adaptable sensor-free system for occupancy detection by employing the recently introduced WiFi fine time measurement (FTM) protocol. FTM measures the round trip time (RTT) of flight between two WiFi-enabled devices. Objects' presence, including the human body in an indoor space, make multi-path reflections that change the measured RTT. My proposed system leverages the variation in the RTT measurements to detect the occupant's presence in an indoor space. This FTM-based occupancy sensing system is able to detect static occupants, and if the room configuration changes, the system can adapt to the new configuration in less than 80 seconds, so this change will not impact the system occupancy detection performance.

Committee:

  • Tariq Iqbal, Committee Chair, (CS, ECE/SEAS/UVA)
  • Brad Campbell, Advisor, (CS, ECE/SEAS/UVA)
  • Rich Nguyen (CS/SEAS/UVA)
  • Seongkook Heo (CS/SEAS/UVA)