Research On Resource Management Algorithm In Broadband Satelite Communication System

With the growing demand for broadband services and the explosive growth of multimedia services, the future of the global Internet should gradually evolve into a variety of heterogeneous network, like the ground Internet,4G mobile communications systems and satellite networks, combined with each other in complementary modes. A major mission of broadband satellite communications is a high rate of information access for multimedia and network applications anytime anywhere. Because the cost of expensive satellite network systems, rocket launchers are expensive and cannot be reused, resulting in valuable satellite bandwidth resources. In order to solve conflicts between the growing demand for satellite broadband and scarce resource, efficient and rational use of satellite network time slot resources, broadband satellite resource management system solutions become one of the hotspots of current research.Based on the research of DVB-S2/RCS broadband satellite communications protocol, this paper focuses on the broadband satellite communications system resource management system, mainly about link layer bandwidth slot allocation strategies and transport layer congestion control algorithm. The main work of this paper is as follows:(1) In-depth study of the characteristics of the satellite channel, typical satellite communications channel models and Markov chain theory, design transfer channel model based on three-state Markov chain model. Then simulate and research the three-state model in DVB-S2/RCS applications. Research results show that the channel model can accurately simulate complex satellite channel, laid the foundation for the study of the upper channel resource management algorithms based on this model system platform.(2) In the study of satellite communication systems DVB-S2/RCS protocols, in particular the bandwidth allocation and scheduling algorithm, because of the multi-scale, self-similarity properties of network traffic, propose wavelet neural network traffic prediction resource management programs. Then simulate and research the wavelet neural network forecasting performance applying to resource management programs. Research results show that the scheme can accurately predict multi-scale long correlation network traffic, and based on traffic prediction algorithm can be a more efficient allocation of bandwidth slots resources, improving system performance.(3) Based on the research of the TCP protocol, against the limitations of using TCP protocols in a long delay, high error satellite links, propose improved cross-layer adaptive TCP algorithm for satellite network. The algorithm mainly by reducing the TCP slow start and congestion avoidance phase of time, the network traffic can peaked faster, and through the link layer queue length to determine the error packet loss is congestion or packet loss. Simulation and the research results show that this scheme can significantly improve the throughput of business, speed up download response time, efficient and rational use of the limited resources of the satellite system.