global-journal-of-computer-science-and-technology-e-network-web-security Global Journal of Computer Science and Technology - E: Network, Web & Security 0975-4350 0975-4172 Global Journals Publishing Group Incorporated 89472 Replication in Mobile Ad-hoc Network using Hopfield Network ChawlaTanu INDIA, The Technological Institute of Textile & Sciences 15 01 2013 13 E15 10 16

Mobile Ad hoc Network (MANET) is a network comprises of mobile nodes connected without any centralized administration. Each node in the network has the ability to share a large number of objects with other nodes. The nodes are connected with other nodes to forward a message to other nodes until the search for node that desire meets. This paper represents a new replication method to recover the performance in distributed system. Objects are replicated on different nodes in the network to minimize the search for an object. It is observed that mobility of nodes predicted using SOM (Self Organizing Maps) technique provides less accuracy. This paper represents an algorithm that uses the network parameters as input condition to replicate objects into the nodes. The decision for each node will be taken by gathering the input conditions and feed to the Hopfield Neural Network. It is view that it will provide more accuracy and improve the performance.

 hop field network mobile ad hoc networks neural networks replication routing routing protocols. Full Text

Mobile Ad hoc Network (MANET) is a network comprises of mobile nodes connected without any centralized administration. Each node in the network has the ability to share a large number of objects with other nodes. The nodes are connected with other nodes to forward a message to other nodes until the search for node that desire meets. This paper represents a new replication method to recover the performance in distributed system. Objects are replicated on different nodes in the network to minimize the search for an object. It is observed that mobility of nodes predicted using SOM (Self Organizing Maps) technique provides less accuracy. This paper represents an algorithm that uses the network parameters as input condition to replicate objects into the nodes. The decision for each node will be taken by gathering the input conditions and feed to the Hopfield Neural Network. It is view that it will provide more accuracy and improve the performance.