Based On Ant Colony Optimization For Qos Constrained Distributed Multicast Routing Algorithm

With the volume of Internet communications business continues to expand, how to find a minimum cost of the network's multicast routing Trees in QoS constraints become a hot research. Research main papers the distributed QoS multicast routing based on ant colony optimization algorithm,which has been solved the problem of routing for a long time and optimize the overall performance of poor and it is easy loop in the existing algorithm when the size of multicast increase.This paper proposes a new Distributed QoS multicast routing mathematical model based on research papers distributed networks, QoS constraints, multicast routing, including the network model, QoS and measurement features and multicast routing model. When the size of multicast increase, we found the problem of the higher mortality Ant and it is easy loop in the existing algorithm. This paper proposes a distributed loop detection with degree constraints algorithm, first of all, the initial establishment of multicast routing tree, multicast routing tree to find and can be set up at the same time to complete. This paper proposes an improved ant algorithm, which define two types of ants agents: forward ant and backward ant which find common bandwidth and delay constraints to meet the minimum cost of the path. Ants have been optimized the initial multicast routing tree on the side of the forward side to make full use of the distribution of the local node, to improve the quality of solutions, reducing the circuit.This paper improves the convergence rules through Information intensity change control,Residual pheromone limiting control and Coefficient of volatile dynamic adaptive to increase the speed of algorithm convergence and improve the global convergence. The impact on the size of the network and multicasting and delay restrictions on the size of the impact test results of show that this algorithm has Adaptability of the large-scale networks and multicast request and relatively short period of time routing, excellent global optimization and convergence of the appropriate time, compared with the BSMA and KMB algorithm.