Edge IoT-Enabled Cyber–Physical Systems with Paper-Based Biosensors and Temporal Convolutional Networks for Real-Time Water Contamination Monitoring

Water pollution poses serious threats to public health and the environment, requiring efficient and scalable monitoring solutions. This paper presents a Cyber–Physical System (CPS) that integrates paper-based biosensors with an Edge IoT architecture and Long-Range Wide Area Network (LoRaWAN) for real-time assessment of water quality. The biosensors detect pollutants such as arsenic, lead, and nitrates with a detection limit of 0.5 ppb. The collected data are transmitted via LoRaWAN to edge devices, where preprocessing and analysis are performed using the Temporal Convolutional Network (TCN) algorithm. The system proposed is compared with existing LSTM systems based on two performance metrics: detection accuracy and latency. Paper-based biosensors are fabricated using silver nanoparticle-functionalized substrates for high sensitivity and low-cost pollutant detection. Data transmission is based on LoRaWAN protocol to have long-range communication with packet loss per cent at a minimum level. TCN algorithm deployment at the edge allows for real-time processing for time-series data analysis due to its high accuracy and low latency properties, compared to LSTM models, which were mainly chosen due to their usage in most applications dealing with time-series-based analysis. Experimentation was carried out by deploying the developed CPS in controlled environments, simulating pollutant levels at different levels and executing them for accuracy in detecting pollutants and the latency of data processing. The system's energy consumption was reduced through efficient edge processing, enhancing the long-term sustainability of its deployments. The TCN framework achieved a detection accuracy of 98.7%, which surpasses LSTM by 92.4%. In addition, TCN reduces latency in processing by 38% to enable fast data analysis and decision-making. LoRaWAN allows for perfect packet transmission of up to 15 km while the loss rate stays as low as 2.1%. These results establish the proposed CPS as reliable, efficient, and scalable for real-time water contamination monitoring. Thus, this research introduces the integration of paper-based biosensors with advanced computational frameworks like TCN and explores its great potential as a transformative development to pave the way toward more sophisticated multi-sensor fusion systems in future studies.