Mobility Prediction

Locating users as they move from one place to another in a cellular computing environment is an essential component in many industrial and networking fields. Tracking mobile users locations, and then, using this data to predict their future locations is the enabler of many modern technologies such as smart cities, automated vehicles, traffic and congestion management, rout recommendations, providing continuous services, migrating services with guaranteed quality of service, low latency communication, etc.

In this direction, mobility predictors, which are designed based on Machine Learning (ML) or statistical models, analyze users' historical location data, extract meaningful information, and learn moving patterns to forecast future locations and trajectories of users' end systems. The goal of research on this thesis topic is to optimise the task of mobility prediction through designing robust, reliable, and accurate predictors while consuming least amount of computational resources.

Related Literature:

[1]: A survey for mobility big data analytics for geolocation prediction

[2]: A new Markov-based mobility prediction scheme for wireless networks with mobile hosts

[3]: Mobility prediction based on machine learning