Trust Based Cross Layer Intrusion Detection System for MANET

Abstract

The MANET (Mobile Ad hoc NETwork) is highly vulnerable and easy to get affected at various levels. The MANET is one of the dynamic network in military sector for wireless communication. To circumvent the existing attacks in MANET this research aims to develop some mechanisms to avoid such attacks. A new method called ANTIDS is developed making use of ANT colony optimization technique as an Intrusion Detection System. The developed system works on Trust values. Trust values are calculated across the nodes. A trust framework is developed which works on IDS engine. The proposed ANT IDS architecture works upon various nodes such us leader nodes, source nodes, neighbor nodes, malicious nodes. We have used the fuzzy logic system to calculate the trust value. Node Selection is based on ANTIDS. This is a heuristic method used for optimization. It s based on the idea on how Ants find the shortest path from their nest to prey. The shortest path of an ANT can be found by using pheromone trails. The ANTIDS algorithm defines the evolution of forward Ant system (FANT) and backward ant system (BANT). A Fuzzy rule set is classified which is used to identify the attacker in the network newline