A clustering hierarchy algorithm based on spectral method and modularity measure(CHSM) is presented in this
paper. The original clustering structure of the networks is given by using the nontrivial eigenvectors, then a parameter
modularity measure is used to evaluate whether the clustering fits for the real networks structure. So a clustering structure
which fits for the real networks can be got by using this strategy. At the same time, the function about the disassortativity
coefficient of energy distributing is presented, and the residual energy of the nodes and the disassortativity coefficient of
energy distributing in the cluster are considered in selecting the cluster head. Simulation results show that the proposed
approach can obtain a more reasonable and steady distribution of clustering, the modularity measure and the disassortativity
coefficient of the clustering are more high, which can prolong the lifetime of networks.