中文摘要：

针对认知无线Mesh网络拓扑结构和可用频谱实时变化的特点，提出一种基于频谱聚合度分簇（SCDC）算法．该算法提出了节点间可用频谱的质量聚合度因子，联合节点位置变化信息，通过计算节点权值实现认知无线Mesh网络分簇的优化．另外，该算法通过簇内成员节点数量的约束阈值实现均衡网络负载．仿真分析证明，SCEC算法在维持网络拓扑相对稳定和提高频谱利用率方面更具优势．

英文摘要：

According to the characteristic of time-varying channels in cognitive wireless Mesh network, a new clustering algorithm based on spectrum convergence degree （SCDC） is proposed. Based on the conver- gent factor of available spectrum quality and node＇s geographical information, the algorithm optimizes the clustering by computing the node＇ s weight. The algorithm designs a constraint threshold of cluster member to balance network load effectively. The simulation results show that SCDC has a good performance on the cost of cluster management and spectrum utilization.