Research Of QoS-aware Scheduling Optimization Technology In Collaborative Mobile Streaming Media System

In recent years, the rapid development of the smartphone industry has witnessed booming prosperity of the mobile Internet. As the most popular Internet terminal, s-martphone has become a major driving force to the growth of Internet users in China and make mobile streaming service promoted to the fifth largest mobile Internet appli-cations. It has become a main entertainment for Internet users to watch the video on smartphone. However, the limited capacity of cellular network, download speed jit-ter and high cost limits the scope of mobile streaming applications. For this, in recent years, academia proposed hybrid network based mobile P2P streaming system. Mobile users can connect with the video severs via B2D connection and connect with nearby users via D2D connection at the same time.Hybrid network based mobile P2P streaming system is mainly divided into two categories:The first category is single-hop broadcasting based mobile P2P streaming system. Based on P2P collaborative technology, users download different parts of the video, then to share them between nodes by broadcasting. The advantage of the system is the high data transfer speed and efficiency. But the scale of the system was limited because users must be in the coverage of communication. The second category is multi-hop routing based mobile P2P streaming system. Users can take advantage of the Ad-Hoc routing to communicate with a user outside the communication coverage. The advantage is that the system can cover more users and resources and the disadvantage is the bandwidth bottlenecks of data transmission and the load of cellular network is not optimal.In multi-hop routing based mobile P2P streaming system, first, the actual topology of segment scheduling link, including the link length and the forwarding nodes’load, determine the transmission performance. Second, some users may download the same segment and the load of each user is not balance. In order to solve the above problems, in this paper, we propose a intermediate node-based segment scheduling optimization strategy and QoS guarantee cellular network load optimization strategy. The following are the main work of this paper:Firstly, we propose a helper node collaboration based scheduling optimization mechanism. We use a weighted directed graph to model of the user and D2D con-nections and reduct the optimazation problem to the shortest path problem. Then we propose a Dijkstra based shortest path algorithm. At last, through a simulation, our algorithm is proved to have a good performance under different system size, different helper node fraction, different helper node cache size and different node density.Secondly, we study the offloading cellular traffic problem of mobile P2P stream-ing system. We model this problem as a minimizing download nodes set problem and prove that it is NP-hard. Based on the model and analysis, we propose a heuristic greedy algorithm to offload system’s cellular traffic and balance each node’s cellu-lar traffic. Through simulation, we find that our algorithm can offload cellular traffic and balance nodes’cellular traffic efficiently.