Feeling the pack: Strategies for an optimal participatory system to sense and recognize noise pollution

Abstract

Noise pollution is one of the many environmental factors that affect modern life. Noisy environments bear upon human brain, by limiting concentration abilities and by impacting the quality of sleep. Some recent works proposed participatory systems, based on common mobile devices, in order to sense and collect noise pollution over urban areas. Such an idea has been drawn on in this work to investigate some strategies for noise-sensing participation in an optimal/less-invasive way, by exploiting clients' adaptation on the strength of server-feedbacks and geographical vicinity among sensors. In particular, we designed, implemented and assessed NoiseHound, a mobile application for Android O.S platforms, which aims at detecting, quantitatively classifying and tracking noise pollution by taking into account the participatory aspects as parameters to improve sensing quality and save local resources such as, primarily, battery power and data exchanges over the Net.