Distributed, Low-cost Particulate Matter Sensing: Scenarios, Challenges, Approaches

The observation and control of particulate matter (PM) pollution in ambient air is increasingly being recognized as an important topic in societies across the globe. Classic measurement approaches provide accurate daily means but are static, expensive and suffer from low spatial resolution and high latency. Distributed measurement grids using real-time capable instrumentation not only can provide spatio-temporally fine-grained readings, but also have the potential to enable novel applications. However, in order for them to be feasible, sufficiently accurate low-cost measurement devices are needed. The noise as well as typically low sensitivity and stability of commercially available low-cost dust sensors result in additional challenges that need to be solved when basing distributed measurement grids on them. This includes signal processing, sufficiently frequent recalibration, suitable data processing and – where applicable – incentivization of users. In this paper, we present different scenarios enabled by distributed PM measurement, discuss the major challenges in such systems and show different approaches that can be used to address them.