Performance Evaluation of Coordinated Node Monitoring Response (CNMR) Based IDS for MANET

摘要

MANET is an infrastructure less network and can work without any devices dependencies. These are short range wireless types of medium which support radio frequency transmission. It has low weight construction of protocol for movable hand held devices so many functionalities is lost which breaches security concerns. Among all of them most threatening is intrusion which deals with various types of attacks affecting the actual process of device. Intrusions can be prevented by using various preventive and measuring components like acknowledgement, monitoring proxy node, behaviour monitoring and change detection, sudden data dropping and modification, infinite delays etc. In the existing mechanism for IDS the mobility and neighbour behaviour entity is not taken for new nodes or some existing misbehaving node. Behaviour of such nodes intimates the network about intruder's activity & hence IDS can be initiated. So identification and removal of intrusion in timely manner is primary task in IDS. However, there is not a deep study of the impact of such attacks on the performance of routing protocols through simulations. The existing static & dynamic routing protocols like ADOV, DSR, OLSR needs to be updated for providing better security against the issues. Thus to achieve proper security in such network of devices it needs up gradation of protocols. Thus, this work focuses its concern on improving the security issues in AODV through improved Intrusion Detection System. This paper proposes and evaluates a novel Coordinated Node Monitoring & Response (CNMR) Based IDS through AACK for AODV protocol. It is a multistep process in which the node & its transmission are in control of some monitoring node. This node can generate responses on the basis of threshold. This quantity of response can be taken as major consideration for identification of intruder's node. Our prime concern by proposing new updates for security is to demonstrate higher intrusion detectio- rates with minimized performance issues. AT the initial level of our research the work is getting better results in comparison with other existing mechanisms.