The mobile ad hoc networks gained its popularity for its various applications, quick and easy deployment. This network does not require any fixed infrastructure for deployment. Mobile nodes in this network communicate with each other through a wireless communication medium and makes this network highly vulnerable to many attacks. One of the well-known attack is packet drop attack. Whenever the network faces any kind of attack suddenly degrades network performance. In this paper, we have designed a protocol known as cross-layer packet drop attack detection using swarm intelligence (CLPDM-SI). This protocol followed a cluster based collective swarm intelligence detection mechanism to find a malicious node in real data acquisition system which undergoes packet drop attack. Our protocol (CLPDM-SI) is compared with a protocol which does not follow swarm intelligence known as Adapting Cross-Layer Approach for Detecting and Segregating Malicious Nodes (ACLDSM) mechanisms. Our comparison is done based on various QoS parameters delay, throughput, packet delivery ratio and false positive detection in the MAC layer. Our result shows significant improvements after incorporating the swarm intelligence. By inspecting CPU and memory utilization time, the algorithm finds the false positive of suspected malicious nodes.
Read full abstract