Abstract
The paper presents the results of a study on the influence of user mobility patterns in the O-RAN (Open Radio Access Net-work) on the parameters of a multi-linear model. The considered multi-linear model is represented as three-dimensional tensors that describe the changes of received signal power or SINR (Signal-to-Interference-plus-Noise Ratio). Factor matrices after the CP (Canonycal Polyadic) decomposition are exploited to separate users by their mobility using the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) algorithm. A model and a set of scenarios are developed for simulating the O-RAN segments and to study the influence of the user mobility on the parameters of the multi-linear model. Simulation results indicate that the number of users in groups with different mobility patterns does not affect the parameters of the multi-linear model and the performance of the DBSCAN algorithm. For the considered scenarios, the best classification results are achieved when the order of the multi-linear model equal to three. Moreover, it is revealed that the topology of a network segment effects on the order of a multi-linear model.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call