Research On QoS Mobile Multicast Routing Algorithm

Mobile network is a combination network of mobile communications and computer networks. Each node has the function of routers and hosts. It has the four characteristics as follows: (1) the dynamic changes of network topology, (2) limited resources, (3) multi-hop communications, (4) the lower security. It meets many challenges for providing QoS (Quality of Service) in mobile network. QoS research has become a hot issue in mobile network. Mobile multicast QoS routing is designed to meet the needs of each routing QoS and simultaneously generate the minimum spanning multicast tree which includes all members of the multicast group.Aiming at the characteristics of mobile network, the QoS parameters are taken into account at ZRP protocol to obtain the largest network topology which meets the QoS requirements. And the ant colony algorithm is explored to solve the optimal multicast tree in the multi-constrained QoS routing problem. When getting the network topology, the improved algorithm can choose a regional agent from each region to transmit data, and provide services for the nodes of the region. The mobile node which has the smallest frequency from the region is selected as the regional agent to construct network model. According to the size of the search path, the improved ant colony algorithm combining with the orthogonal statistical methods, choose the suitable orthogonal table which determines the impact of various degrees of QoS parameters to determine pheromone volatility. The impact of various degrees of QoS parameters makes most of ants tend to select the impact of degree routing. Furthermore, in the search process, the probability of selection methods is used. Two groups of ants search separately and interact pheromone after each search to avoid the search process into the local minimum solution. Combined with multiple QoS constraints, the optimal path is selected from the two selected paths. When choosing the next node, each ant imports the probability of selection that has been introduced in the last to accelerate the search speed. Experiments show that the algorithm has better performance. It can set up the QoS multicast tree quickly and efficiently, convergence to the global optimal solution fast, and reduce the reconfiguration frequency of multicast tree due to node mobility effectively.