Research On The Resource Optimization Based On Network Coding In Multi-source Optical Multicast Networks

With the rapid development of broadband networks and new multicast businesses, the high quality of multicast services is the requirement of more and more users. At the same time, the high reliability requirement of multicast businesses transmission quality has been improved. These requirements have brought the huge challenge to the traditional electric domain IP multicast. Optical network has the advantages of wide bandwidth and high speed. Therefore, it becomes the ideal way of the high rate of multicast transmission and high bandwidth businesses. Network coding technique can improve the reliability and throughput of network. The technology is introduced into the optical multicast networks to fully take the advantage of network coding and optical network, and also able to provide an effective solution for broadband and realize the reliable multicast services.With the emergence of new business, multi-source optical multicast is being applied more and more widely. So the research of the multi-source optical multicast is meaningful and necessary. There exist some problems when the conventional optical multicast routing algorithm is applied in multi-source optical multicast, such as the poor network resources utilization, and cannot maximize the network throughput. This thesis not only analyzes several typical intelligent optimization algorithms and their application in optical multicast routing optimization, but also analyzes multi-source optical multicast routing optimization based on network coding in chapter 2.Network coding that used in optical multicast can bring benefit obviously, but resource overhead in the optical network has been increased. Intelligent optimization algorithm can properly solve the NP hard problems because of its superiority, such as the optimization of network coding. The existing algorithm is easy to fall into local optimum with search space increasing. Number of code links can be minimized according to genetic algorithm in chapter 3. Algorithm can guarantee the diversity of population and avoid falling into local optimal by designing difference between the maximum cross. Adaptive crossover probability guarantee the diversity of population at the beginning and ensure stability when individual in a population is generally well, when the solution falling into the local optimum. The simulation results show that the proposed algorithm is more efficient and have faster convergence speed, which can take a very short time to find the minimum numbers of coding link solutions compared with the contrast algorithm.With different kinds of network interconnection, network scale expanding and diversification of client processing ability, it is inevitable that the heterogeneous problem increasingly prominent. Layered multicast technology can well solve the problem of heterogeneous networks. The multi-rate layered optical multicast algorithm based on minimum coding network has proposed in chapter 4. The allocation of layer rate can use the trade-off layer rate algorithm the proposed. It can maximize the network throughput through determining the layer rate from destination nodes which have the same receiving capacity. Minimum coding network is put forward to realize the reuse of link, and the link resources that saved are used to transmit data for the layer behind. The simulation results show that the proposed algorithm can improve network throughput, optimized network throughput.