Network bottlenecks in OLSR based ad-hoc networks

Abstract

All nefarious players must be able to maximize profit on an attack. In the context of communication networks, part of that need translates to finding the best topological location for the attack. While in static networks this is an easily solvable problem, in mobile ad-hoc networks (manet) having constant topology fluctuations, this becomes a formidable task. Highly important asset nodes are those having numerous paths traversing through them, with bottlenecks being nodes that all data must pass through. Once such nodes are compromised, the damage can be catastrophic. Hence, network defenders must also find these nodes so they can be better protected.In this paper we propose a method that ensures finding such bottleneck nodes (assuming one exists) in olsr based manet, with linear cost. This is accomplished without adding overhead on the network. We experiment and compare, using network simulation tools, multiple types of attackers over a diverse set of topologies.