IoT-based Framework for Real-Time Air Quality Health Index Monitoring using LoRaWAN and Low-cost Sensors: A Proof of Concept for Toronto

Abstract

It is widely recognized that air quality directly influences human health, which has driven growing research interest in developing real‑time index-based systems for air quality monitoring, often leveraging wireless sensor networks. This work presents an Internet‑of‑Things (IoT) implementation of an Air Quality Health Index (AQHI) framework aimed at assessing environmental conditions at fine spatial scales that leverage low-cost parts, low energy demand, and thus provide cost-effective opportunities for wide-area or dense deployment for monitoring with a network of wirelessly connected sensors. Compared with conventional air quality stations, which are typically bulky, costly, and deliver measurements with low spatial detail, the IoT approach offers inexpensive hardware, a small form factor, straightforward installation, and the ability to build scalable networks that provide fine‑grained data for localized analysis. Our system tracks concentrations of key airborne pollutants, such as ozone (O₃), nitrogen dioxide (NO₂), and fine particulate matter (PM_2.5), which serve as indicators of overall air quality and are already incorporated into the Canadian reporting system. An exclusive web‑based platform was developed to enable live, near real‑time monitoring, long‑term data archiving, analytical processing, and the visualization of air quality information through an intuitive, user‑friendly dashboard. The system computes the AQHI for each sensor deployed which can provide near real time reporting, monitoring, and with known sensor locations, the data could be mapped. Interaction with this scalable platform is through either a computer, tablet, or even smartphone if an Internet connection is available. A single microcontroller enables sensor‑derived gas and PM measurements to be transmitted over the Internet to user dashboards. This study outlines the complete hardware configuration and software framework underpinning the proposed system. A case study for multiple sensor deployments to compute AQHI is presented along with an evaluation of cost and energy consumption benefits of using this system.