This paper argues about the deployment positions of Network-based Intrusion Detection System and suggests the "Distributed Network Security Sensors" distributed among the nodes of the internal network to monitor traffic. We study the tradeoff between cost and monitoring coverage to determine the positions and processing rates of the sensors. To handle the uncertain nature of flow, we build fuzzy expected value optimization models and develop a hybrid intelligent algorithm to obtain the deployment strategy. From the experiments in actual and synthesized network topologies, we observe that a small number of low-speed sensors are sufficient to maintain a high monitoring coverage. It also depicts that deploying DSS is much more efficient in larger topologies.