Fourteenth International Conference on The Constructed Environment

Abstract

A bespoke multi-sensor building occupancy system was installed at Queensland University of Technology in Brisbane, Australia. The system was designed to localise occupants within a space and obtain information on the internal air quality. The World Health Organisation defines air pollution as the single biggest environmental threat to human health, with indoor air pollution a major concern. Presently people spend most of their time indoors and not enough is understood about how to improve indoor air quality. Ventilation can be an effective solution for removing pollutants, however its effectiveness depends greatly on environmental and architectural conditions. Knowledge of occupancy level and the distribution of pollutants indoors can facilitate effective ventilation design. A distributed occupancy and air quality sensing system was designed, which comprised infrared, CO2, temperature and relative humidity sensors and microphones. The system was installed in an office and measured during normal working hours. Data was collected over a period of several weeks and statistical analysis was conducted on the measurement data. The effectiveness of the system in classifying occupancy patterns was judged against a source of ground truth. Distributed CO2 sensing proved to be an effective, low-cost strategy for localising occupants within a room. Obtaining occupancy state by using voice activity detection algorithms on audio data presented several challenges, including privacy preservation, computing power required, and data storage. Several alterations and improvements are proposed for a subsequent sensing system to be installed in a UK location.