Abstract: Mobile ad-hoc networks are temporary wireless networks. Network resources are abnormally consumed by intruders. Intrusion detection techniques are used for the network attack detection process. The receiver collision, low transmission power and misbehave report authentication attacks are generated in the routing process. The acknowledgement scheme is used to control the routing based attacks. Service level based attacks are raised using service requests. Classification methods are used to discover the service level based attacks. The nodes are grouped into clusters. The leader nodes are assigned for the clusters. Cluster Dependant Leader Election (CDLE) and Cluster Independent Leader Election (CILE) schemes are used in the system. The system optimizes the leader election and intrusion detection process. The system uses the Enhanced Adaptive ACKnowledgment (EAACK) scheme for routing based attack detection process. RSA and Secure Hash Algorithms are used to secure the EAACK scheme. The Bayesian Classification algorithm is used for service level attack detection process. Node mobility is managed by the system.

Keywords: Mobile ad-hoc networks, Intrusion detection, CDLE, CILE, EAACK, Bayesian classification algorithm.