Titolo della tesi: Innovative Sensing and Analytics for Aquaculture and Underwater Environmental Monitoring
This thesis explores innovative data analytics and real-time monitoring solutions for aquaculture and underwater environmental systems. Using bio-loggers deployed on Atlantic salmon and advanced sensor networks, this research investigates stress detection in fish and the integrity of underwater sensor data in real-world environments. The work is divided into two primary areas: stress moni- toring in aquaculture and environmental monitoring for submerged ecosystems. In aquaculture, the deployment of bio-loggers allowed the capture of high-resolution data on fish behavior and physi- ological parameters. These findings, coupled with environmental data, enabled real-time detection of stress and provided insights into how fish respond to human interventions, such as crowding and feeding practices. In the environmental domain, this research addressed the critical challenges of sensor drift and fault detection in underwater networks, proposing data-driven approaches to ensure long-term data reliability. Additionally, a forecasting metamodel was developed to predict near-future environ- mental conditions, aiding in the anticipation of critical events and optimizing sensor operations. By combining real-world data with advanced analytics, this thesis contributes to the enhancement of both aquaculture management and underwater environmental monitoring, demonstrating the broader implications for industrial applications such as aquaculture, marine tourism, and submerged cultural heritage preservation.