A Pay-and-Stay model for tackling intruders in hybrid wireless mesh networks

• Published in: Simulation (San Diego, Calif.) Calif.), 2013-05, Vol.89 (5), p.616-634
• Main Authors: Pathan, Al-Sakib Khan, Abduallah, Wafaa Mustafa, Khanam, Shapla, Saleem, Habibullah Yusuf
• Format: Article
• Language: English
• Subjects: Backbone; Clients; Finite element method; Handling; Intrusion; Materials handling; Networks; Wireless networks
• ISSN: 0037-5497; 1741-3133
• DOI: 10.1177/0037549712467827

Intrusion handling in wireless mesh networks (WMNs) is a relatively less addressed topic. The difficulty may lie in the fact that there are other wireless networks for which some intrusion detection or prevention schemes are proposed that could also be applied in some way in a WMN setting. As those schemes are contributing, researchers may not find it necessary to specifically focus on this field. Another critical reason may be the difficulty in developing an effective scheme for WMNs. In fact, the structural differences among various wireless ad hoc networking technologies make it imperative to devise the mechanisms in subtle but critically different ways. For WMNs, there is a proper network backbone that is called mesh backbone (which is not present in many other wireless network counterparts), which supports the fringe part or the mesh clients. Hence, it is often possible to install the intrusion handling mechanisms or agents in the stable part and allowing some flexibility in the client or fringe parts. Nonetheless, instead of thinking in this pattern, we take a different approach of tackling intrusion by allowing an intruder to stay in the network as long as it proves to be worthy of staying in the network by supporting the network’s regular activities. The idea is that by not always directly purging rogue entities, one could exploit the intruder’s resources, thereby benefiting the network. We call our approach an intrusion-tackling mechanism and term it a Pay-and-Stay model. Alongside presenting the details and analysis of our model, in this paper, we also present the basics of various forms of intrusion handling in such types of networks. Using our evaluation results, we found that the model could be very effective in handling intruders and defending the network against a broad range of security attacks.