A multi-attribute decision based routing selection ( MADRS) algorithm in IPv6 mobile networks was pro-posed to avoid overloading problems caused by the concentration of a large quantity of data in a few mobile routers. The multi-attribute decision model was established by introducing multiple decision attribute information including the bandwidth, the packet loss rate and the number of accessed nodes, as well as using the combination weighting approach to calculate the weight of each attribute of decision. This algorithm takes into account the load distribution state of the mobile network and the type of node data comprehensively, and allocates the data traffic reasonably and thereby balances the network load effectively. This improves the performance of the entire mobile network. Simula-tion results show that the proposed algorithm can control network flow, and shorten the network delay by about 50%. This algorithm has significantly improved the whole network performance with a lower cost and has good a-daptability to the network environment.%针对IPv6多宿移动网络容易出现数据流集中于少数移动路由器节点，从而导致负载过重问题，提出了一种基于多属性决策的负载均衡路由选择算法。通过引入移动路由器节点的带宽、丢包率、挂载节点个数等多个决策属性信息，使用组合赋权法计算各决策属性的权重，建立了多属性决策模型，移动节点依据该决策模型选择最优的移动路由器节点来转发数据。该算法综合考虑移动网络负载分布状况及节点数据类型，合理分配数据流量，可均衡全网络流量。仿真实验表明，该算法使移动网络流量得到控制，且网络延时降低约50％，网络整体性能得到了明显的改善，同时该算法代价小且具有良好的网络环境适应性。
展开▼
