Modeling And Performance Analysis Of Multimedia Streaming Service Quality Assurance Based On Software Defined Network

With the rapid development of the network technologies and the explosive growth of the demand for real-time services like multimedia service,the current network is no longer provide a good quality of experience(QoE)for multimedia service users.Software defined network(SDN)is rising in response to this dilemma.SDN brings a huge innovation by applying the idea of decoupling the control panel and forward panel.This decoupled design makes it possible to extend the network by just replacing the underlying universal switching equipment.SDN also allows each layer of the network to evolve independently.Thus,SDN can largely reduce the cost of network extending and evolving.SDN is also very stable because it can reduce the complexity and the vulnerability of the network.However,even under the SDN environment,the rapid growth of real-time multimedia service will still make a real-time multimedia user wait in line.This will inevitably cause a high delay which can compromise QoE greatly.Both the academics and the industry are paying a lot of attention to solve these problems.Nowadays most of the researches on quality of streams under SDN environment are focusing on routing optimization.These studies usually provides a better QoS guarantee for multimedia services by applying better routing algorithms.However,researches focusing on dealing with the various real-time multimedia services with traffic bursts and high concurrency can hardly be found.This work takes all these challenges into consideration and applies network calculus and the concept of admission control to obtain a better QoS guarantee.The main contribution of this thsis is as follows.Firstly,to deal with the high concurrency of video services,the admission control mechanism is used for parallel admission to multimedia streams.With the advantages of admission control in preventing network congestion,the admission mechanism can greatly improve the efficiency of network operation and reduce the probability of network congestion.Secondly,an aggregation scheduling model is proposed to aggregating the video streams after the admission stage,and network calculus is applied to compare the aggregated flows.Compared with the existing hybrid scheduling model,the proposed aggregation scheduling model can effectively reduce the backlog and delay of multimedia services.This aggregation scheduling model can provide better quality guarantee for multimedia users.Thirdly,the correctness and validity of the proposed model are verified by theoretical derivation and OMNET ++ simulation.