Development and Experimental Validation of an IoT-enabled Water Quality Measurement Framework

Abstract

Sensors act as transducers that convert physical signals into electrical outputs for analysis. When integrated with IoT, they enhance water quality monitoring by improving accuracy, autonomy, and real-time detection of parameters such as pathogens, temperature, and Total Dissolved Solids/Electrical Conductivity (TDS/EC). In line with SDG 6, this study developed an IoT-based system for on-site monitoring and real-time data transmission to relevant authorities. The system uses TDS/EC sensors to measure dissolved ion concentration and a DS18B20 digital temperature sensor to provide temperature compensation for conductivity values. Data are processed via an Arduino Mega and transmitted to the Thing Speak cloud. Our results show temperature stability (28–30°C) and demonstrate that compensation significantly reduces variability, stabilizing TDS/EC readings and improving correlation. The compensation results demonstrated that raw TDS/EC values had a weak correlation with temperature, whereas compensated readings remained stable across the same conditions. The temperature histogram highlights the narrow range of environmental variation, reinforcing dataset stability. Histograms of raw versus compensated data confirm that DS18B20-based compensation enhances measurement accuracy and reliability, ensuring that detected anomalies reflect true water quality variations rather than environmental noise. Collectively, these plots validate that DS18B20-based compensation significantly improves accuracy and reliability in real-time IoT water quality monitoring. In summary, the correlation and regression analysis clearly demonstrate the effectiveness of the temperature compensation mechanism. A strong positive correlation (r = 0.82) and high R² value (0.67) indicate that raw TDS/EC measurements are heavily influenced by temperature fluctuations. After applying DS18B20-based temperature compensation, these effects were significantly reduced, with the correlation dropping to 0.18 and R² decreasing to 0.03. This confirms that temperature compensation enhances the stability, reliability, and interpretability of conductivity-based water quality measurements.