Machine Learning Optimization in Dual-Function Meta-IoT Sensors for ISAC

The integration of the Internet of Things (IoT) with meta-materials advances both communication and sensing technologies. Using Meta-IoT sensors, environmental data can be acquired by analyzing the frequency response from reflected signals. In this study, we introduce a dual-purpose Meta-IoT sensor capable of simultaneously sensing environmental conditions and enabling communication. Machine learning (ML) plays a key role in optimizing sensor design and power management within these meta-material IoT systems. Hence this work leverages machine learning (ML) to optimize sensor design and power management in Meta-material IoT (Meta-IoT) systems. Specifically, Gaussian process regression (GPR) is used to explore and optimize sensor structures, while reinforcement learning adapts power management strategies dynamically. Simulation results highlight that the ML-driven methods enable adaptable and efficient Meta-IoT systems.