提出了一种基于层次分析法的超级节点选取策略。该策略刻画了不同用户对不同应用的需求度量,根据阈值法过滤部分不满足最低要求者得到备选集合,在此基础上构建层次分析模型,给出相应的选择结果。仿真实验表明,该机制能使用户满意度达到95%以上,同时整体网络在检索延迟、网络带宽消耗等有比较好的折衷。
A super node selection strategy based on analytic hierarchy process (AHP) is proposed. In this strategy, the network parameters threshold filter is set up based on user-demand metrics calculation, the filter to the existing super nodes is applied to get the preliminary nodes set; and then analytic hierarchy process (AHP) model is constructed. The simulation results show that the node selection strategy can improve the user satisfaction degree to above 95%, and achieve a good balance between network searching delay and network bandwidth consumption.