Yan, Si2020-06-162020-06-162020-06-162020-05-22http://hdl.handle.net/10012/15998The fifth generation (5G) networks are expected to accommodate various applications with diversified quality-of-service (QoS) requirements. To this end, some new technologies are employed in 5G networks such as software-de ned networking (SDN), network function virtualization (NFV) and network slicing. SDN emerges as a promising architecture towards flexible and agile network operation. By decoupling control functions from substrate nodes, SDN enables informed network control and function recon figuration on programmable switches. NFV partitions the network functions from the dedicated hardware platforms which enables on-demand deployment of network functions. Thanks to the employment of these new technologies, fine-grained and customized in-network control mechanisms can be realized to improve network performance. In this thesis, we develop a transmission protocol for video-on-demand (VoD) streaming in the SDN/NFV-based 5G core network. By exploiting the flexibility of scalable video coding and the caching resources in the core network, we develop a novel selective caching policy for in-network congestion resolution. An enhanced transmission policy is proposed to improve the throughput and the network resource utilization by sending cached video packets once the congestion event is over. The proposed protocol is able to adapt to traffic dynamics and varying network environment, and it is shown to effectively alleviate network congestion with balanced throughput and resource utilization. Simulation results are presented to validate the efficiency of the proposed protocol in terms of packet delay, goodput ratio, throughput and resource utilization.encommunication networkingMaster Thesis