Wireless sensor network (WSN) plays an important role in the field of IOT (Internet of things), which performs the function of information perception. Thousands of devices as well as sensors are spread in specific areas to collect all kinds of physical information to pass onto the Internet. However, the data gethered from sensors' interfaces is real-time, extremely large and unstructured, hence requiring great effort in mapping to the conceptual application layer. To customize and develop IOT systems more efficiently, this paper proposes an approach based on runtime model to managing wireless sensor networks. First, manageability of wireless sensors is abstracted as runtime models which automatically and immediately propagate any observable runtime changes of target resources to corresponding architecture models. Second, a composite model of wireless sensors is constructed through merging their runtime models in order to manage different kinds of devices in a unified way. Third, a customized model is constructed according to the personalized management requirement and the synchronization between the customized model and the composite model is ensured through model transformation. Thus, all the management tasks can be carried through executing operating programs on the customized model. The feasibility and efficiency of the approach are validated through a real case study of smart community.