A Novel Approach for User Equipment Indoor/Outdoor Classification in Mobile Networks

The ability to locate users and estimate traffic in mobile networks is still one of the major challenges when it comes planning optimizing networks. Since indoor location not always possible or precise, having distinguish from outdoor can be a valuable alternative and/or improvement. In this paper, two different machine learning algorithms are presented classify user’s environment, whether outdoor, using only data Long Term Evolution (LTE) network. To test both algorithms, measurement campaigns were done. Both used smartphone gather side. first campaign was done across 6 cities, ranging small rural areas large urban environments, while second on city. On campaign, Network Traces (NT) also collected network algorithm consists Random Forest (RF) relies Short Memory (LSTM), thus covering more traditional deep approaches. results varied 0.75 0.91 F1-Score, depending validation strategy, showing promising results.