Data-Driven Construction of User Utility Functions from Radio Connection Traces in LTE

In recent years, the number of services in mobile networks has increased exponentially. This increase forced operators to change their network management processes ensure an adequate Quality Experience (QoE). A key component QoE is availability a precise model for every service that reflects impact performance variations on end-user experience. this work, automatic method presented deriving Quality-of-Service (QoS) thresholds analytical models several from radio connection traces collected Long Term Evolution (LTE) network. Such QoS reflect minimum below which user gives up its connection. The proposed relies fact experience influences traffic volume requested by users. Method assessment performed with real taken live LTE networks. Results confirm packet delay or throughput are critical factors analyzed services.