由于无线传感器网络中节点部署密度大,因此数据采集存在较高冗余度。基于模糊聚类理论,提出了一种利用格贴近度来建立模糊相似矩阵,进行聚类后依据在尽可能少的观测节点数目条件下使得信息量最大的准则,遴选出冗余节点。然后针对这部分冗余节点给出了特定的休眠调度机制以节省能量。实验结果表明,提出的策略能够在保证整个传感系统的有效性和精确性的基础上,显著延长网络生命周期。 Due to the large node deployment density in wireless sensor networks, there is high redundancy in data collection. Based on the theory of fuzzy clustering, this paper proposes a criterion of using lattice closeness to establish fuzzy similarity matrix and clustering to make the maximum amount of information under the minimum number of observing nodes, and selects redundant nodes. Then a specific sleep scheduling mechanism is given for this part of redundant nodes to save energy. The experimental results show that the proposed strategy can significantly extend the life cycle of the network while ensuring the validity and accuracy of the entire sensing system.