Research On Technology Of Load-Balance Placement Of Gateways In Wireless Mesh Network Based On Immune Model

Abstract:In a Wireless Mesh Networks (WMNs), the load-balance of gateways would be a bottleneck for network performance. Thus strategically placing and connecting the gateways to the wired backbone is critical to the management and efficient operation of a WMNs.In this paper, we are around the gateways’load balancing problem to do research, the gateways’load-balance deployment issues and solutions were proposed, which consist in getting the two goals of minimum gate-ways and load balance in gateways placement such that QoS requirements are satisfied. The following tasks have been accomplished in this thesis:1. To solve the problem of the gateways’load-balance deployment problem, the Hungry_Placement algorithm based on the hungry value of gateways was proposed, to reach the two goals of minimum gate-ways and load balance in gateways placement.The paper defined gateways’hunger value to measure load-balance of gateways, and proposed the Hungry_Placement algorithm. When we assign network nodes for each cluster, we should make the hungry value of gateway become close to their average of hungry value as much as possible and achieve load balance placement of gateways in the end, at same times, it always meets the QoS constraints during the entire clustering process.2. To further optimize gateways’load balancing deployment issues, the paper built immune model for wireless Mesh network, and proposed the IMGA_Placement algorithm based on the immune model, to better realize the optimization of the load-balance deployment of gateways.Through simulating biological immune system and using the affinity and immune memory mechanism of immune theory, we built the immune model, proposed the IMGA_Placement algorithm, also used affinity and concentration to measure, not only assessed the pros and cons of the extent of the antibody for the antigen, also assessed the degree of similarity between the antibody and antibody concentration. These could directly reflect the quality of the antibody itself good or bad, and antibody population diversity.3. To solve the problem of gateways’load imbalance caused by the the changes in the distribution of network traffic after the gateways were deployed, the strategy of clustering dynamically adjusted based on the the immune model was proposed, in order to achieve the maximum extent of gateways’load-balance at the entire network operation phase.Using the memory mechanism of immune model to strengthen local search or inhibit premature convergence, so that the algorithm quickly converged to the global optimal solution, and real-time assessed gateways’load-balance at the entire network operation phase. The Hunger-algorithm and IMGA_Placement algorithm dynamically adjusted network clustering program, fast and efficient solved the problem of gateways’load imbalance caused by the the changes in the distribution of network traffic quickly and efficiently.