IoT Driven Air Quality Surveillance and Forecasting

In order to lessen dependency on expensive sensors, this study introduces a low-cost model that makes use of computing power at the edge and the Internet of Things in an effort to develop affordable air quality monitoring and predictive platforms. The Raspberry Pi, also known as a device that provides edges with computing power, applies the Kalman Filter (KF) method to improve low-cost measurements by 27% at the boundary of the device. By combining data with related mistakes, the methodology uses the KF algorithm to predict the level of six pollutants in the air, including SO2, NO2, and PM2.5, with lightning speed. The KF procedure's higher predictive ability is shown by comparisons with three traditional prediction techniques: Autonomous Integrated Moving Standard, Increasingly Weighted Move Average, and Simple Moving Average, among others. The root-mean-square error measurement showed a significant decrease of 68.3% on averaged. These results highlight the system's potential to promote intelligent farming and are consistent with its potential for implementation in several domains where precise, immediate time quality of air monitoring is essential.