基于谱分析与模块度,提出一种无线传感器网络分簇算法(CHSM),首先利用非平凡特征向量获得传感器网络的原始簇结构;然后借助模块度的增量来评估、合并原始簇,从而形成一个与真实网络相匹配的簇结构;同时设计了一种能量异配度函数,并利用各节点的能量异配度及其剩余能量在各个簇内选取簇头节点,仿真结果表明,CHSM算法找到的簇结构具有更高的模块度,其选取的簇头节点具有更高的能量异配度,进而表明了所提出的算法能有效延长网络的寿命。 Based on spectral analysis and module degree, a wireless sensor network clustering algorithm (CHSM) is proposed. Firstly, the original cluster structure of sensor network is obtained by using non-trivial eigenvectors. Then, the original cluster is evaluated and merged by increment of module degree, A cluster structure matching with the real network is proposed. At the same time, an energy heteroscedasticity function is designed, and the cluster head node is selected in each cluster by using the energy isomorphism of each node and its residual energy. The simulation results show that CHSM algorithm finds The clustering structure has a higher module degree, and the selected cluster head node has a higher degree of energy mismatch, and then shows that the proposed algorithm can effectively extend the life of the network.