Detecting Critical Nodes for MANET Intrusion Detection Systems
SECPERU '06 Proceedings of the Second International Workshop on Security, Privacy and Trust in Pervasive and Ubiquitous Computing
| Hi-index | 0.00 |
Mobile ad hoc network (MANET) routing algorithms have been the focus of a great deal of research over the last few years. This research has primarily focused on the performance, energy efficiency, quality of service (QoS), trust, and security issues of these algorithms. Given the cost and complexity of performing repeatable field experiments, most researchers naturally turn to network simulators to validate their theories. Network simulators allow researchers to quickly and cost-effectively analyze the impact of new approaches under various network traffic and mobility scenarios. Simulation results, however, must be vetted in real-world conditions using commercially available hardware and software to discover the shortcomings of modeling assumptions. Real-world testing, however, can be very expensive and time-consuming and beyond the budgetary constraints of most researchers. In response, we have extended our MANET emulator testbed, mLab [1], to include a collection of host-based kernel network traffic monitoring modules and visualization utilities. In this paper, we present the implementation details of these host-based kernel modules and provide illustrative experimental results to demonstrate how they can be used in conjunction with mLab to capture performance metrics using repeatable and user-defined mobility and traffic scenarios. The illustrative examples are based on NIST's implementation of the AODV and DYMO ad hoc routing protocols [2,3]