Application Layer Multicast Streaming Media Key Technology Research And Implementation

As the need for media application of the concurrent users increases gradually, the combination of the streaming media technology and ALM solves the problem of the one-to-many communication, and has become an important research direction on the current Internet. This thesis makes research and implementation on the key technology in the ALM video living system—streaming media processing which is based on the project of CNGI(China Next Generation Internet).The streaming media processing has become the focus research because of its importance and complexity. Through the analysis of key technology in the stream media processing, a design scheme which is for streaming media processing in ALM system is presented, and its rationality and efficiency is demonstrated. On the base of that, the streaming media processing system constructed on the IPv4/IPv6 network with the object oriented technology is implemented in this thesis.The thesis discusses the design and development process of the system in detail, including collecting/encoder, data encapsulation/cache management, network transmitting, network transferring and playing. With the synchronization mechanism and scheduling tactics, this thesis makes all the phases cooperate with each other. Also the paper takes a deep study on the key problems in each phase and solves them, such as interface design, network transmission control, cache replacement strategy,the scheduling of the data package, the speed control of the writing and reading packets. Dynamic changes of Internet bring out the process of stream media unpredictable problems, with the exception-safe and exception-neutral mechanism, the system robustness is increased, also the problem of judgment nodes leaving is solved in the video living broadcast system which reduces the nodes re-connection time.The system is executed on the CERNET and CNGI-CERNET2. The test results and analytical reports for this subsystem are provided in this thesis, and the experimental data shows that the streaming media schemes used in the subsystem is reasonable, effective, and the system is scalable, low-cost, easy to implement, wide application, etc.