<![CDATA[Water pollution is a serious environmental issue in Indonesia, necessitating an accurate and real-time water quality monitoring system. This study developed an Internet of Things (IoT)-based monitoring system using an ESP32 microcontroller integrated with four analog sensors, namely pH, Dissolved Oxygen (DO), temperature (DS18B20), and Total Dissolved Solids (TDS). The system was designed with a layered architecture comprising a sensing layer, edge processing, a communication layer based on the MQTT protocol on a Linux Ubuntu server with a Mosquitto broker, and an application and storage layer using a MySQL database and a web interface based on Laravel with real-time visualization using JavaScript and CSS. Sensor calibration was performed using a multipoint calibration approach with linear and polynomial regression, accompanied by temperature compensation and digital filtering (median filter and Exponential Moving Average). Performance evaluation was conducted through 30 simultaneous measurements against standard laboratory instruments, resulting in an average system accuracy of 90.9% with R² values ranging from 0.806 to 0.956. The TDS and temperature parameters showed the best accuracy at 94.2% and 93.3%, respectively, whereas the pH and DO achieved 86.8% and 89.3%, respectively. The results of this study indicate that the developed ESP32-based IoT system is capable of generating reliable data (accuracy above 90%) for water quality monitoring in tropical environments. This research provides a practical contribution in the form of a solution that supports the implementation of sustainable monitoring in accordance with Government Regulation Number 22 of 2021 and Sustainable Development Goal (SDG) 6.]]>
Copyrights © 2026
