Network-driven handover in 5G

• Autor(a) principal: Lopes, Ana Rita Moreira
• Data de Publicação: 2020
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10362/144489
• topic: 5G; New Radio; Handover; Real Time Events; Quality of Experience (QoE); User Equipment (UE); Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

Currently, users’ expectations regarding technological performance are constantly increasing. An example of this is the growing consumption of multimedia content via the Internet. Multimedia applications with a variable number of users/requests have variable demand over time that may expose the limitation of the network channels. This may cause a problem of demand mobility generated by the service/application. Each generation of mobile networks has specific handover processes, which in the case of 4G can be controlled according to the applications requirements, with the possibility of multiconnectivity. This process was massified in 5G. The main contribution of this dissertation is the development and analysis of decision models for controlling the video streaming and user association to a BS in the network architecture. The scenario considered refers to a football stadium with multiple points of view – video streams – that each spectator can request to view on their cell phone or tablet. The developed simulator models the stadium scenario using a combination of services, which occur on the 5G network. Vertical handover generated by the network is used,aidedbynetworkslicing. Thenetworkslicingactsinthepartofthebandwidthdivision between the different antennas and allows the throughput of the different broadcast (FeMBMS)channelsto becontrolledbytheservice -theradionetworkcapacitylimitsthe throughput. The results obtained in a case of 80000 spectators who select different beams over time, considering8basestations(BS),showthatthequalityofexperienceishighonlywhenthe handover and the control of beam diffusion by BS are managed according to the application requirements. The network recovers from huge peaks by handling as many requests at once as possible. Instead of the user only getting the steam in a good quality or not getting it at all, the network performs a best-effort solution of downgrading the quality of multicasting in order to expend less resources with the same quantity of requests. The network state is taken into consideration. Although there are load peaks on the network, it is never congested.