AI-Enhanced Rainfall Retrieval Using Commercial Microwave Links in 6G-IoT Networks: Advances, Challenges and Opportunities

Escalating global climate change has intensified the urgent demand for high-precision, real-time rainfall monitoring. While rainfall retrieval based on commercial microwave links (CMLs) currently faces key challenges such as data scarcity and insufficient model generalization. The integration of core Sixth-Generation (6G) technologies within the Internet of Things (IoT) framework presents promising solutions. These emerging technologies offer inherent advantages, including wide coverage, low cost, easy deployment, and strong real-time capability. Building upon these strengths, it has significantly enhanced CMLs rainfall retrieval performance by integrating 6G’s terahertz communication and ultra-low latency transmission, IoT-enabled edge intelligence, Integrated Sensing and Communication (ISAC), and Artificial Intelligence (AI) for data processing and real-time analysis, the performance of CML-based rainfall retrieval has been significantly enhanced. This provides a viable technical pathway for achieving sub-kilometer accuracy meteorological monitoring and building multi-dimensional sensing systems for smart cities. The main works of this paper are as follows: (1) A comprehensive review of recent advances in CML-based rainfall retrieval, covering both model-driven and data-driven retrieval technical routes from fundamental principles; (2) Designing and conducting a rainfall retrieval experiment using a long short-term memory (LSTM) model based on real-world E-band CMLs measurements from Nanjing, China, which robustly verifies the high-precision retrieval potential of data-driven approaches in complex urban propagation environments. (3) An in-depth analysis of emerging challenges, such as high-frequency data fusion, faced by CMLs rainfall retrieval technology in scenarios deeply integrated with 6G-IoT scenarios, along with discussion of potential technical solutions.