Mobility Prediction Using Non-Parametric Bayesian Model

Predicting the future location of users in wireless networks has numerous important applications, and could in turn help service providers to improve the quality of service perceived by their clients (e.g. by applying smart data prefetching and resource allocation strategies). Many location predictors have been devised over the last decade. The predictors proposed so far estimate the next location of a specific user by inspecting the past individual trajectories of this user. As a consequence, when the training data collected for a given user is limited, the resulting prediction may become inaccurate. In this paper, we develop cluster-aided predictors that exploit the data (i.e., past trajectories) collected from all users to predict the next location of a given user. These predictors rely on clustering techniques and extract from the training data similarities among the mobility patterns of the various users to improve the prediction accuracy. Specifically, we present CAB (Cluster-Aided Bayesian), a clusteraided predictor whose design is based on recent non-parametric bayesian statistical tools. CAB is robust and adaptive in the sense that it exploits similarities in users’ mobility only if such similarities are really present in the training data. We analytically prove the consistency of the predictions provided by CAB, and investigate its performance using two large-scale mobility datasets (corresponding to a WiFi and a cellular network, respectively). CAB significantly outperforms existing predictors, and in particular those that only exploit individual past trajectories to estimate users’ next location.