Research Of Time-Shifting Technology Based On P2P Real-Time Streaming Media

P2P real-time streaming media technology is full use of node resources in network, alleviate pressure on the server, expand the scale of system services, and reflect the real-time characteristic. Therefore, this kind of technology becomes a hot research in streaming media technology area. Times-shifting as the key technology of streaming media has been widely used in non-real-time streaming media system, but not effectively used in real-time streaming media. Time-shifting operation allows the user to select anywhere in stream time to play. However, the current existence of real-time streaming media system can not perform efficiently in time-shifting function, due to node caching algorithm and system structure in system. Therefore, improving caching algorithms, designing of system architecture, and then efficiently combining two parts to realize the time-shifting function has important theoretical and practical significance.This paper commences studies according to the key issues of time-shifting technology in real-time streaming media as followed:For the defects of current tree, mesh, and DHT structure in real-time streaming media system, this paper proposed effective Semi-distributed P2P time-shifting streaming media structure-SPTS. SPTS structure using dual-ID improves the node's perception of space and resources to address the logical network and physical network mismatch problem of the DHT structure and increase the efficiency of resource discovery; using the advantages of centralized topology and fully distributed unstructured topology solves the problem of high nodes load, large consumption of resources and maintenance overhead in tree and mesh structure. Simulation results show that, SPTS structure can shorten the transmission path, reduce the search delay, and maintenance costs.For the problem of lack of first video blocks and insufficient hot video blocks in current real-time streaming media system, this paper proposed IPPI, LSPI cache algorithm applied in P2P streaming media systems. IPPI, LSPI algorithm divided cache area into two parts: Distributed Stream Buffer (DSB) and Playback Buffer (PB). DSB stores real-time video blocks which is used to share with other peers, and remains unchanged. PB stores video blocks which are used to play and changes according to the progress of players. IPPI and LSPI algorithms solved the problem of lack of the first video blocks and insufficient hot video blocks. Simulation results show that IPPI, LSPI algorithm effectively reduce the server stress and the loss rate of the video blocks.This paper using the research results of real-time structure and cache algorithm designed a P2P-based real time streaming media system model SPTSS. SPTSS system combining IPPI, LSPI caching algorithm and the SPTS structure can effectively perform time-shifting function; reduce the pressure on server and nodes. The test results show that in the process of playing, the server and nodes pressure values were maintained at a very small range.