<![CDATA[Various technologies have been developed to monitor dissolved oxygen (DO) and turbidity in aquaculture, yet integrated evaluations focusing on biofloc systems, particularly those involving IoT and Machine Learning (ML), remain limited. This review analyzed 32 studies published between 2020 and 2026 using the PRISMA 2020 framework to examine DO measurement, turbidity measurement, IoT integration, and ML applications in biofloc aquaculture. To support methodological discussion, several studies from broader aquaculture and water-quality monitoring contexts were also considered. The reviewed literatures shows that IoT-based and manual methods are the most commonly used approaches for DO and turbidity monitoring. IoT systems, mainly based on ESP32, ESP8266, and Arduino platforms, support real-time monitoring and automation. ML models such as Random Forest, LSTM, and CNN-LSTM are frequently applied for water-quality prediction, anomaly detection, and decision support. However, challenges related to sensor calibration, data availability, and model generalization remain. These findings suggest a growing shift toward more intelligent and integrated aquaculture monitoring systems.]]>
Copyrights © 2026
