Detecting and Preventing Selfish Node Presence in Mobile ADHOC Network Using Watchdog with Fuzzy Bayesian Filter Approach

Abstract

                Secured data transmission in mobile network is most concerned research issue focused by various researchers. Particularly network with the presence of selfish nodes would lead to tremendous degradation in network performance. Selfish node uses the resource only for its own purpose not for their neighbours so it is very important to detect the selfish node to improve the network performance of MANET. This is rectified in the proposed research work by introducing the method namely Authentication and Monitoring based Cooperative Attack Detection (AMCAD). In this work, initially selection of cluster head is done by considering the various factors such as node degree and trust level of nodes along with the energy and bandwidth. Here cluster head selection is done by using improved ACO algorithm where VIKOR method is introduced for the pheromone update process and authentication is performed by introducing the modified group signature algorithm. Here initially random values generated by both cluster head and the cluster members will be exchanged with each other. By using that random number, signature generation will be done. This method enables all nodes involved in the environment to perform authentication by using those random values based generated signatures. After node authentication, data communication is initialized where optimal route path selection is performed with the concern of selfish node presence. Here the selfish nodes are avoided by integrating the fuzzy logic based Bayesian watchdog. The overall implementation of the These techniques are applied in a wide range with respect to the IDS which is centrally deployed or distributed in nature. This IDS is called as Distributed IDS (DIDS) when it is used in distributed environment like distributed WSN or any ad-hoc type network  security level.