<![CDATA[Air pollution in Jambi City, due to the increasing PM2.5, often exceeds the national quality standards. This situation requires a uniform monitoring solution, since the number of government reference stations is limited. An IoT-based monitoring system using low-cost sensors can be an alternative, which are flexible and real-time monitoring. We developed an IoT-based monitoring system using two low-cost sensors (PMS5003 and PMS7003), which were collocated with and validated against the BAM-1020 reference instrument operated by the Meteorology, Climatology, and Geophysics Agency (BMKG) at the Sultan Thaha Jambi Climatology Station. This quantitative study aims to evaluate the accuracy of the two sensors against the reference instrument for the period December 2025 to January 2026. The analysis was carried out using the statistical parameters MBE, MAE, RMSE, the correlation coefficient (r), and the coefficient of determination (R²), followed by a linear regression-based correction of the sensor readings. The results showed that both sensors were able to follow the pattern of changes in PM2.5 concentration quite well (R2 = 0.60 for PMS5003 and 0.59 for PMS7003), although they tended to produce higher values (overestimate). The PMS7003 sensor showed slightly better accuracy with a lower error rate (MBE=10.46; MAE=12.53; RMSE=18.02 µg/m3) compared to the PMS5003 (MBE=11.84; MAE=13.60; RMSE=19.60 µg/m3). This overestimation is consistent with the influence of high relative humidity and the technical limitation of light-scattering sensors, which cannot distinguish solid particles from water droplets, as reported in previous studies. Overall, both sensors are reliable for real-time monitoring, but still require further calibration.]]>
