Predicting a User's Next Cell With Supervised Learning Based on Channel States

TL;DR

This paper presents a machine learning system that predicts a user's next cell in cellular networks using channel state information and handover history, enabling better resource management and location-aware services.

Contribution

It introduces a novel classification-based prediction system employing SVMs with a pre-processing structure, using readily available network data.

Findings

High prediction accuracy in simulation scenarios

Timely predictions suitable for real-time applications

Effective use of CSI and handover history for classification

Abstract

Knowing a user's next cell allows more efficient resource allocation and enables new location-aware services. To anticipate the cell a user will hand-over to, we introduce a new machine learning based prediction system. Therein, we formulate the prediction as a classification problem based on information that is readily available in cellular networks. Using only Channel State Information (CSI) and handover history, we perform classification by embedding Support Vector Machines (SVMs) into an efficient pre-processing structure. Simulation results from a Manhattan Grid scenario and from a realistic radio map of downtown Frankfurt show that our system provides timely prediction at high accuracy.