Big data analytics for dynamic network slicing in 5G and beyond with dynamic user preferences

Network slices represent a valuable technique to utilize the full resources of fifth-generation (5G) platforms. This allows verticals to control and utilize separate virtual systems on the above identical physical framework. In this article, we suggest a self-sustaining network slices (SNS) structure that combines self-learning, self-slicing control efficiency optimization, and self-management regarding system facilities for achieving an adaptable control approach under unanticipated system circumstances. The forthcoming version of NodeB (gNodeB) layer splitting and packet schedule layer slicing, while networking level slices are the three layers that the suggested SNS paradigm decays the SNS command into such levels. At the networking stage, every gNodeB has access to network services over a long period with a coarse resolution of capabilities. Additionally, we employ a transfer-learning strategy to switch from a model-driven system to an autonomous, self-improving SNS monitoring. The suggested SNS architecture is intended to improve new applications' QoS efficiency significantly.