AirQ-IoT: A Real-Time Indoor Air Quality Monitoring and Multi-Hazard Risk Management Framework for Low-Cost Edge-Cloud Architectures
Abstract
Indoor air pollution drives a substantial share of respiratory, allergic and thermal-comfort morbidity in residential buildings, yet continuous multi-hazard monitoring remains rare in low-cost Internet-of-Things deployments, which typically report a single composite pollution index and ignore the multidimensional nature of indoor environmental risk. This paper presents AirQ-IoT, an end-to-end open-source framework that integrates (i) an ESP32 multi-sensor edge node, (ii) a Node.js/Express/MySQL ingestion backend with WebSocket broadcast, (iii) a Python FastAPI analytics service executing 94-feature engineered pipeline and (iv) a real-time web dashboard. The framework contributes: a 94-feature engineered pipeline with cross-sensor disagreement and interaction-ratio features that improve forecasting accuracy; five complementary health-risk indices (composite IAQI, ASHRAE 55 PMV/PPD thermal comfort, mold germination risk, dust allergy risk and skin/mucous membrane irritation) computed in parallel rather than collapsed into a single score; a 4-layer anomaly detection engine that combines threshold rules, rate-of-change spikes, multi-sensor disagreement and statistical z-score detection; and a reproducible 60-day empirical evaluation across three residential room archetypes representative of subtropical climate. The complete platform, dataset and analysis scripts are released as an open-source artifact to facilitate further research and deployment.
