A review on node activity detection, selfish & malicious behavioral patterns using watchdog algorithm

In the existing system, the node may sometimes behave selfishly. To act selfishly means that nod sometimes is unwilling to forward the packets to the further node as prescribed in route from source to destination in order to save their resources and this may seriously affect the performance of the network. In the proposed system, the network is distributed into clusters and every cluster has their own cluster head that helps the nodes to transmit and receive the packets more prominently. Every node in the cluster assigns themselves with the set of friends and enemies. Basically, the cluster is a combination of both nodes (i.e., selfish nodes and normal nodes). As described above, every node has a set of friend nodes and enemy nodes so if the node is selfish node then it will transmit the packet to the friend list node. Selfish node doesn't send a packet to the enemy nodes and this shows that selfish node always takes the greedy decisions. However, the detection of selfish node done by the watchdog process and sometimes it may fail in the complex network, by generating false negative and false positive decision and that leads in performing wrong operations. To make the process more complex we add up another group of nodes known as malicious nodes. Malicious nodes by nature will accept a packet from all the nodes but transmit those packet to wrong nodes that leads to losing of resources of many nodes in the network. Malicious node sometimes alters the data in the packet or corrupt the data. To identify the behavior pattern of a node by two neighborhood node we thought of applying chord algorithm.