Development of an IoT -based fishpond water monitoring system using Arduino with enhanced depth measurement accuracy

Water quality monitoring technology is an indispensable in fishery aquaculture. This paper is focuses on the deep pond aquaculture. The experimental study developed a fishpond monitoring system based on the Internet of Things. The system monitors the real-time water quality parameters such as depth, water temperature, air temperature, and humidity. Then the data is transmitted and stored through ONENET middleware and local database. Finally, the real-time monitoring results and alarms are displayed on a mobile application. The accuracy of depth measurement was tested in both clear and turbid water environments, with the average error rate of 2.84% and 4.09%, respectively. Moreover, Kalman filter-BP neural network algorithm is used to enhance predictive accuracy and computational efficiency. Results shown that the root mean square error of the Kalman filter-BP neural network algorithm is optimized to 5.95%, compared to the standard BP neural network.