Abstract:
Occupancy detection and measurement provide vital information for smart buildings to adapt to human occupants. A cost-effective way to gather accurate and up-to-date occupancy information, while preserving the privacy of the inhabitants of the building is yet to be widely adopted. Pyroelectric infrared (PIR) motion detectors have been extensively used as an occupancy detection device for building automation and control. This research aims to review PIR motion detectors and explore their applications for occupancy detection and measurement in smart office buildings and residential homes. In this thesis, the proposed occupancy monitoring system has two main components: hardware system implementation and data processing. The hardware implementation describes the design of a low-cost, low-maintenance, wireless motion detector system using PIR sensors and unmodified Fresnel lenses. This hardware system was then used for developing and testing the data processing methods. In one of the proposed data processing methods, an array of PIR motion detectors was used to track a single person. Simulations and experiments were carried out to evaluate the performance of the tracking system. The results have shown that the human tracking system can effectively track a single target with accuracy approximately equals to ¼ of the distance between the PIR sensors, thus allowing a relatively small number of sensors to be used. In the other proposed data processing method, we attempted to use hidden Markov model (HMM) to model and measure occupancy. The results show that with the right combination of motion detectors and data inference method, the PIR based system is a viable strategy for the development of reliable occupancy detection systems in enclosed and open-plan office buildings. Simulation and experimental results demonstrated that the accuracy of the single sensor method is up to 97% and the multiple sensor method is up to 99.7%. The robustness of the model was tested, and a method for repeated training was described to adapt to the changing behaviour of the occupants.