Classifying users based on their mobility behaviour in LTE networks
Faculty of Sciences. Mathematics and Computer Science
S.l. , 2014
Tenth International Conference on Wireless and Mobile Communications (ICWMC 2014), June 2014
University of Antwerp
When, in cellular networks like Long Term Evolution (LTE), there is a dense deployment of cells and/or users move at high velocities, handovers will occur frequently. This will have a severe impact on the Quality-of-Service (QoS) experienced by the users as there will be frequent call drops and a low throughput due to a service interruption time that is relatively long compared to the cell stay time. In order to mitigate these problems, users that experience a so called high mobility should be steered appropriately (e.g., to a cell on which they can be camped for a longer time). The first prerequisite for steering users that experience high mobility is identifying these users and predicting their future behaviour. In this paper, we present an algorithm that identifies users that follow similar trajectories through a cell. To perform this identification, a modified version of the Dynamic Time Warping (DTW) algorithm is used. Results show that the developed algorithm is able to detect users which follow similar trajectories adequately and is also able to distinguish between users that follow different trajectories through a cell.