Real-time and predictive analytics of air quality with IoT system: a review

Abstract

Environmental pollution particularly due to the emission of combustible gas from industry, haze, and vehicles, that has always been a major concern. Continuous monitoring of the air quality is hence essential to ensure early precaution or preventive measure can be taken in eliminating potential health risk which may be done via Smart Environmental Monitoring system with the Internet of Things (IoT), which is cost-effective and efficient way to control air pollution and curb climate change, IoT applications along with Machine Learning(ML) can make the data prediction in real-time. ML can be used to predict the previous and current data obtained by sensors. This review describes the existence of an integrated research field in the development of the environmental monitoring system and ML method. The findings of this review interestingly show that (i) various communication module is used for environmental monitoring system. (ii) Very less integration of IoT together with predictive analytics, it is separately to study for air pollution monitoring system. (iv) Data analytics for Air Pollution Index (API) prediction along with IoT, with various communication protocols can assist in the development of real-time, and continuous high precision environmental monitoring systems. (v) Machine Learning (ML) Regression algorithm is suitable for prediction and classification of concentration gas pollutant, while ANN and SVM algorithm is used for forecasting.