基于特征矢量中心性的分布式分簇算法

Distributed clustering algorithm based on eigenvector centrality

针对无线传感器网络分簇结构优化和负载均衡性问题,提出一种基于特征矢量中心性的分布式分簇算法。通过计算特征矢量中心性分数值确定全网络节点的重要性,将当前影响最大的节点作为簇头,筛选簇头的连通子集作为簇成员,逐次迭代执行,实现网络簇划分。通过选择加入网关节点保证簇头子集连通性,对其它边缘散落节点提出新的入簇策略,优化整个分簇结构。仿真结果表明,该分簇算法能分布式均匀地划分簇,优化了分簇数目,网络的连通子集保证了簇结构的稳定性,改善了网络负载均衡性能。

A distributed clustering algorithm based on eigenvector centrality was proposed for clustering optimization and load balancing of wireless sensor networks.The importance of the whole network node was considered using the eigenvector centrality score value,and the node with the current largest influencing node was used as the cluster head,the connected subset of the cluster head was taken as the cluster member by screening,network cluster partitioning was implemented by successive iterations.Through choosing to join the gateway node,the connectivity of cluster head was ensured,and a new clustering strategy was proposed for other edge scattered nodes to optimize the whole clustering structure.The simulation results show that the proposed clustering algorithm can distributedly divide clusters evenly and optimize the number of clusters.Connected subset of the network can guarantee the stability of the cluster structure and improve the network load balancing performance.