Cost Optimization For Linear Network Coding About Multisource Network

As we all know, network communication has become an important role of modern society. With the introduction of network coding theory, it completely changed the knowledge of data transmission and greatly increased the maximum multicast rate of the network. The research on network coding mainly focuses on the design of network coding scheme,and how to construct a specific and effective coding scheme. However, in order to make the network coding more practical value, we have to have to go further to consider the network coding optimization problem in the research results now. Although there already have a lot of research on single source network coding, but because of the multi-source network is not a single source of heterogeneity of the simple superposition of information, so the current multi-source network research is still in its infancy. But also for the multi-source network coding optimization problem is a big challenge. In this paper, we propose a suitable continuous-flow network coding model for a specific multi-source network with no delay. And the algorithm based on solving this kind of optimization problem is proposed. The simulation results show that the proposed algorithm can be realized.In this paper, we first consider a new method to construct the linear network coding for the network resource minimization problem in multi-source network topology. Based on our proposed linear network combinatorial model, we propose a nonconvex optimization problems with constraints. Then for this kind of very complex problem, we can only be transformed into approximate convex optimization problem, and then use the penalty function to eliminate constraints, and finally solve it through the unconstrained algorithm, or the first penalty function to eliminate constraints, and then the Genetic Algorithm is used to combine the algorithm to find the optimal solution. As a contrast,we can also use the dual theory to further get the lower bound of the global optimal solution of the problem. Finally, based on the proposed algorithm, we simulate the corresponding multi-source network topology, design simulation experiments, and analyze the feasibility of the model and the convergence of the algorithm. The experimental results verify the effectiveness of our proposed linear network coding scheme and the minimization of the cost of a network resource can be achieved by this algorithm.