No abstract

Wireless Mesh Networks (WMNs) do not rely on any centralized administration and they are built by the connection of various static and mobile entities (i.e. nodes). The cooperation and coordination between these network entities is very essential to establish a secure routing path. Such distributed nature of WMNs increases the vulnerability of routing protocols. Modeling and computing trust in such a distributed environment is essential since the network is self-organizing, self-configuring and relies on multihop routing. Therefore, to ensure a secure route discovery and its maintenance, it is important to compute trustworthiness of individual nodes in a cooperative manner for discovering neighbors, selecting routers and announcing topology information in WMNs. In order to detect trustworthy nodes in the networks, we propose a model based on Multiple Criteria Decision Making (MCDM) which quantifies node behaviors into discrete quantity. The proposed scheme ensures detection of malicious and misbehaving nodes in the network which is verified through code simulation.

Wireless Mesh Networks (WMNs) consist of static router backbone which provide connectivity and coverage to mobile clients. They can also be viewed as a special case of wireless multi-hop Mobile Ad-hoc Networks (MANETs). The IETF group as well as the IEEE 802.11s working group have recommended applicability of several existing MANETs routing protocols for WMNs. However, in WMNs, client to client communications are through parent gateway via static routers, while in MANETs every arbitrary node pairs can act as source or destination. Therefore, it is of practical importance to re-evaluate MANETs routing protocols and make suitable modifications so that the protocols perform efficiently in WMNs environment. In this paper, we propose E-AODV, which is a variant of traditional Ad-hoc On-demand Distant Vector (AODV) routing, a protocol developed for MANETs. To evaluate E-AODV's performance and suitability in WMNs, a comparison of the protocol with traditional AODV and M-OLSR (a modified OLSR for WMNs), is carried out.

No abstract