Spatial and temporal mobility analysis in LTE mobile network

As smartphones and location-based services gain mainstream popularity, human location histories analysis and user's future location prediction attract increased interest. Predicting the human mobility mainly on users' spatial movement has been conducted in extensive studies. In this paper, we focus on the user mobility prediction from spatial-temporal perspective, and implement several slot-based continuous next-place prediction models considering various contexts. The entropy is also measured to explore the influence of the intrinsic of user's trace on prediction performance. We experiment with real-world data derived from the factual LTE mobile network, and evaluate the performance of proposed predictors with several metrics. By thoroughly experimental analysis, our finding shows that user mobility is highly dependent on both temporal and spatial behavior. Models considering the spatial context of current and previous locations can achieve high accuracy and robustness in both spatial and temporal prediction. Also, the spatial movement independent temporal predictors gain an edge for users whose mobility is in a regular pattern.