International Journal of Engineering and Technology

Adhoc network is a collection of nodes that are capable to form dynamically a temporary network without the support of any centralized fixed infrastructure. There is no central controller to determine the reliable & secure communication paths in Mobile Adhoc network. Each node in the Adhoc network has to rely on each other in order to forward packets, thus highly cooperative nodes are required to ensure that the initiated data transmission process does not fail. In a mobile Adhoc network (MANET) where security is a crucial issue and they are forced to rely on the neighbour node, trust plays an important role that could improve the number of successful data transmission. Larger the number of trusted nodes, higher successful data communication process rates could be expected. In this paper, a model is proposed for Intrusion Detection System and then Worm-Hole attack is assumed in the network, statistics are collected to design intrusion detection engine for MANET Intrusion Detection System (IDS). Important features extraction and rule inductions are applied to classify the data set. Training of data set and on the basis of confidence value generated in training is used for testing of data set and investigated the accuracy of detection engine by using support vector machine. In this paper True Positive generated by the detection engine is very high and this is a novel and statistics based approach in the area of Mobile Adhoc Intrusion detection system. Though this research is carried out for Adhoc network environment but this is also applicable to Wireless Sensor network