Intrusion Detection for Mobile Ad Hoc Networks Based on Node Reputation

Abstract

The mobile ad hoc network (MANET) is more vulnerable to attacks than traditional networks, due to the high mobility of nodes, the weakness of transmission media and the absence of central node. To overcome the vulnerability, this paper mainly studies the way to detect selfish nodes in the MANET, and thus prevent network intrusion. Specifically, a data-driven reputation evaluation model was proposed to detect selfish nodes using a new reputation mechanism. The mechanism consists of a monitoring module, a reputation evaluation module, penalty module and a response module. The MANET integrated with our reputation mechanism was compared with the traditional MANET through simulation. The results show that the addition of reputation mechanism can suppress the selfish behavior of network nodes and enhance network security.