Research On Key P2P Technology Based On Streaming Media Live System

With the further development of multimedia transmission technology,streaming media live show is expected to become the k iller application of the Internet today.In a large-scale live scenario,the traditional client/server model is not able to meet the system’s performance requirements any more.Peer-to-Peer(P2P)model has the advantages of high scalability,robustness and low cost.Each node has the same status in the network,and shares its own storage,computing,bandwidth and other resources with other nodes.Which not only reduces the pressure on the server,but also avoids a single point of failure,it is very suitable for large-scale streaming media live system.As a real-time application,the streaming media live system is very sensitive to the transmission delay and the playback continuity,because these two factors directly affect the quality of experience,how to meet the requirements of low transmission delay and playback continuity is a very important research direction.This paper introduces the development history,characteristics and research status of P2P technology,and expounds the theoretical knowledge re lated to streaming media.This paper analyzes and studies the key technologies about existing network construction method and data scheduling strategy.O n this basis,a new P2P streaming media system scheme is proposed.The scheme includes P2P network construction method based on upload bandwidth and biased data scheduling Strategy.The main work of this paper is as follows:First of all,this paper studies the existing mainstream streaming media system and classic P2P transmission protocol,analyzes the salient features,advantages,disadvantages and the applicable scope of each scheme,and draws valuable experience from them.Summed up the impact of system transmission delay and fluency o f the two important factors: network structure and data scheduling,and focus to explore on these two dimensions.Secondly,this paper starts from the system topology,introduces the characteristics of structured P2P network and unstructured P2P network.Based on the characteristics of the two structures and the upload bandwidth,a hybrid hierarchical network construction method is put forward,in which nodes with strong service capability are closer to the content source node,thereby speeding up the distribution of streaming media data.On the basis of that,a biased decentralized scheduling strategy is raised.It sets the priorities separately according to the degree of urgency of the data and the service capability of the nodes to ensure the fluency and load balancing.Finally,the proposed P2P scheme is simulated on the simulation platform.The concrete steps include the construction of the experimental platform,the realization of the scheme and the analysis of the experimental results.The main factors that affect the performance of streaming media live system,such as startup delay,transmission delay,packet loss rate,and compared with the existing schemes.A large number of experimental data shows that the proposed solution achieves the desired purpose,it reduces streaming media transmission delay,improves the quality of experience,and it’s extraordinary suitable for large-scale streaming media broadcast system.