Detection of Precipitation Based on the Received Signal Level of Commercial Microwave Links and GRU Neural Networks

In the recent years, signals from commercial microwave links have been used not only their main purpose, communication, but also for detection of precipitation which affects the link. Among different techniques for processing the measurements on the commercial links, AI based techniques have been developed. The aim of this paper is to investigate precipitation detection techniques based on commercial microwave link measurements and gated recurrent unit (GRU) neural networks.