The A TG Interference Source Localization based on Machine Learning

The extensive deployment of Air-to-Ground (ATG) system significantly exacerbates interference issues with terrestrial 5G NR system, thereby elevating the importance of localization interference originating from A TG system. While existing method can locate interference sources to a certain extent, they predominantly require additional hardware installations or infrastructure modifications, escalating both costs and complexity. Addressing this challenge, this research introduces an innovative framework that utilizes Signal-to-Interference-plus-Noise Ratio (SINR) measurements collected from operational networks. These SINR values are geographically encoded, and ResN et is used to extract features from these encoded maps and locate the interference sources. Furthermore, the study examines the impact of varying the number of sensing base stations on the accuracy of interference source localization. Experimental results demonstrate that this approach substantially enhances the accuracy and reliability of interference source localization, offering significant practical benefits for intelligent interference management and 5G network optimization.