Real-Time IoT-Enabled Smart Water Quality Monitoring System Using Embedded Multi-Parameter Sensor Analytics and Attention-Based Predictive Modeling for Environmental Protection

Abstract

Rapid industrialization and urbanization have led to a significant deterioration in water quality, posing serious threats to both the environment and public health. Conventional water quality assessment, which relies on manual sample collection and laboratory analysis, is time-consuming, labor-intensive, and incapable of providing real-time information. To address these limitations, this study presents an IoTbased Smart Water Quality Monitoring System (SWQMS) that leverages embedded sensor analytics. The system deploys conductivity, dissolved oxygen, turbidity, and pH sensors to acquire real-time data and identify anomalies through edge processing on microcontrollers. IoT communication protocols transmit the processed data to a cloud platform, where it can be stored, visualized, and subjected to predictive analysis. Field deployment of the SWQMS in freshwater sources yielded measurements that exhibited a high correlation with laboratory reference values, thereby demonstrating the system's accuracy in monitoring water quality and detecting pollution events in a timely manner. The proposed SWQMS continuously measures pH (7.03–7.14), turbidity (3.2–3.6 NTU), dissolved oxygen (6.85–7.10 mg/L), and temperature (24.9–25.6 °C) using built-in multi-parameter sensors and edge analytics. IoT-based data acquisition, real-time cloud visualization, and predictive forecasting enable the early recognition of anomalies and the delivery of automated warnings, thereby supporting proactive water quality management in ponds, industrial wastewater treatment facilities, and aquaculture operations.