Balancing High-Load Scenarios with Next Cell Predictions and Mobility Pattern Recognition

摘要

Knowing where a mobile user will be next candeliver a tremendous increase in network performance underhigh load, as this knowledge enables pro-active load balancing.To derive this information, sequences of traversed cells are fedinto pattern detection algorithms. After the training phase thelearned model predicts each user’s next cell. Even for complexscenarios, the prediction accuracy can exceed 90%. Predictionsare used to rearrange mobile connections in a simulated high-load scenario centered around an event at a soccer stadium.To prevent call drops for mobile users targeting the stadium,apropriate resources in the predicted next cell are reserved. Theresults exceed 20% in improvements for throughput and calldrop rates, enabling the network to bear a much higher loadbefore stalling.