In this project, we are developing new sensor networking tools that monitor indoor air quality and allow a user to obtain information about what events cause changes in their air quality.
Fig. 1. The system architecture consists of three components: the home sensor, air server, and smart device. The home sensor is a Dylos particle monitor modified to include a Beaglebone Black with WiFi access, SHT21 temperature and humidity sensor, and a local SQLite database. The air server contains a PostgreSQL database, which stores data from all deployed air monitors and analyzes data based on the amount of particulate matter, temperature, humidity, and time. The server's analysis finds patterns in activities that change the air quality and provides the user with a real time graph of the data. The smart device (e.g., smartphone or tablet) runs our Android app, which alerts the user when there are changes in the air quality and asks for feedback about what event may have just caused that event. The app learns from the training and later learns what class of event has just occurred based on sensor data.