IoT-Based Air Quality Monitoring System for NO2 and O3 Concentrations Using Neural Networks in the City of Ibarra

Abstract

Air quality monitoring is an essential component of urban environmental management; however, medium-sized cities such as Ibarra often lack updated measurements, limiting evidence-based decision-making. This study presents the design and implementation of an Internet of Things (IoT) system for monitoring nitrogen dioxide (NO2) and ozone (O3), pollutants closely associated with vehicular traffic and atmospheric degradation. The system integrates calibrated electrochemical sensors, a data acquisition and wireless transmission module, and a web-based platform for real-time data visualization. To improve signal interpretation, a multilayer artificial neural network was implemented and trained over 71 epochs, achieving a classification accuracy of 97.55% on the independent test dataset for three-class air quality level classification. Field measurements conducted at locations with high vehicular flow reported NO2 concentrations between 0.00 and 0.03 ppm (approximately 0–57 µg/m3) and O3 concentrations ranging from 15 to 33 ppb (approximately 30–65 µg/m3), remaining below World Health Organization guideline limits. The results indicate that the proposed solution constitutes a low-cost and scalable tool for preliminary urban air quality assessment and real-time classification, particularly useful for municipal environmental agencies, researchers, urban planners, and public health stakeholders in intermediate cities with limited monitoring infrastructure.