The wireless sensor network (WSN) has emerged as a promising technology. In WSNs, sensor nodes are distributedly deployed to collect interesting information from the environment. Because of the mission of WSNs, most node-wide as well as network-wide activities are manifested in packet traffic. As a result, packet traffic becomes a good data source for modeling sensor node as well as sensor network behaviors. In this article, the methodology of modeling node and network behavior profiles using packet traffic is exemplified. In addition, node as well as network anomalies are shown to be detectable by monitoring the evolution of node/network behavior profiles.