Personal air pollution exposure assessment using wearable sensors

Air quality have become a global issue with increasing attention. There is a growing concern in the public about both in- and outdoor air quality. In order to monitor individual air pollutants exposure in real-time, we developed several wearable air quality monitoring devices to study personal exposure to different pollutants in different environments. The devices are equipped with different type of sensors to measure NO₂, aerosols, CO₂, etc, in addition to environmental parameters sensor to measure temperature, relative humidity and pressure. In order to optimize the accuracy, we compared different retrieval approaches such as multiple linear regression, generalized linear model, neural network, etc. This allows us to perform the personal exposure study with a high temporal resolution in the order of seconds. We classified daily activities into different categories: different ways of commuting such as bus, tram, subway, bicycle, on foot; indoor activities like cooking, lighting candles, etc.; outdoor exercises next to busy street, in a park, etc. In this presentation, we will present our first result of this personal exposure study.