Detection of false alarm in handling of selfish nodes in MANET with congestion control

In a mobile ad hoc network, the mobile nodes will have the characteristics of mobility and constraints in resources. Since, the mobility is high, the nodes may move randomly and fastly, which lead to network partitioning. The resource constraints leads to a big problem as decrease in performance and the network partitioning leads to poor data accessibility. To improve the data accessibility, we have proposed several data replication techniques. Most of the users at different places assume that mobile nodes co-operate fully in terms of sharing their memory space. But In reality, some nodes may decide as not to co-operate with others or partially co-operate with other nodes. The behavior of these selfish nodes leads to decrease in over all data accessibility of the network. We have explored the impression of selfish nodes in a MANET from the perspective of replica allocation and developed selfish node detection algorithm that considers the partial selfish node and fully selfish node as selfish replica allocation. The replica will be allocated using specific SCF tree concept. An alarm will be raised based on the selfish behavior of overall nodes called overall selfishness alarm. But the alarm will also be initiated because of network disconnections too but it seems and treated as overall selfishness alarm, it will affect the overall performance of the network. The concept of the paper deals with detection of false alarm as differentiated from overall selfishness alarm and to inform the other nodes at route as exactly where the disconnections occur to select the next best alternative path and also to increase the performance with increased congestion control. Detection of attacker node in the network and should be informed to all others in the network.