Design of a Decentralized and Predictive Real-Time Framework for Air Pollution Spikes Monitoring

Abstract

Exposure to air pollution spikes cause health problems to regularly exposed organisms, raising the need to monitor them in real-time. Existing air pollution monitors mainly use a cloud-centric design considering relatively constant pollution, therefore duty-cycling sensors with long sleep periods to save their batteries. Such design is however inefficient for monitoring pollution spikes. Furthermore, since spikes vanish rapidly, integrity of monitoring data is very important. This paper presents a framework integrating edge-centric design and blockchain in monitoring air pollution spikes, while using short-term prediction artificial intelligence to timely alert pollution emitters about exceeding long-term average pollution limits defined by standards.