Supervised Machine Learning in 5G NSA Networks for Coverage and Performance Analysis

5G NSA networks are now distributed in many nations all over the world. However, telecom operators have limited ways, like Drive Test, to analysis the network’s performance. In this situation, we will introduce a supervised machine learning based approach to predict the 5G NSA network coverage and performance in a specific area. Our solutions will use the 4G MDT data as the eigenvalue. We firstly use the ML algorithms to analysis the accuracy between 4G MDT data and 4G DT data. Then, select the most important index by analyzing the impacts of 5G NSA coverage and performance. Finally, we will get the rules by training the supervised machine learning model between 4G MDT data and 5G NSA coverage and performance outcomes. The results indicate that the supervised machine learning based approach works well. Telecom operators can use it for the further development of 5G networks.