User Mobility Dataset for 5G Networks Based on GPS Geolocation

Abstract

Geolocation technology is the most exciting area of advancement in 5G, leveraging massive sources of accurate location data to provide users with effective location-positioning services and applications. As research on user mobility prediction is steadily growing in the context of 5G networks, the need for available mobility-related data is of utmost importance to support the development and evaluation of new individual mobility patterns. This paper presents a novel mobility dataset generation method for 5G networks based on users' GPS trajectory data. First, we propose aggregating the user's GPS trajectories and modeling his location history by a mobility graph representing the set of cell base stations he passed through. Second, we implement the proposed modeling approach to build a custom mobility dataset and provide a detailed description of our methodology. The generated dataset relies on mobility traces from the real-world Geolife dataset and contains the mobility graph records of 128 users. Finally, we discuss selected use cases for which we believe our dataset would be valuable.