Analysis and Impact of Trust and Recommendation in Social Network based Algorithm for Delay Tolerant Network

Abstract

Message delivery in sparse Mobile Ad hoc Networks (MANETs) is difficult because of the network is not connected properly. To provide path for message in a delay tolerant network is an area of concern in last few years and many researchers have introduced so many techniques for routing. In delay tolerant sparse network number of nodes in the network is very fewer and probability of losing the data is very high. Many approaches have been proposed in recent past but most of them failed to perform significantly. Social studies are a successful approach for information propagation in wireless networks. Bridge nodes are identified on their centrality i.e., to broker information exchange among other disconnected nodes. Because of complexity in centrality metrics the concept of ego networks is developed in which nodes are not required to exchange information about the entire network, but only locally available information is given. Then SimBet Routing proposed the exchange of pre-estimated „betweenness‟ centrality metrics and locally determined social „similarity‟ to the destination node with combination with the human opinion as a recommendation. An opinion dynamics based approach is proposed in this work which calculates the trust value of a node based on the opinion of other nodes in the network. SimBet Routing outperforms PRoPHET Routing, particularly when the sending and receiving nodes have low connectivity. The influence of social network analysis or impact of social network analysis is very high while packet dissemination in delay tolerant network but none of the reported work shows any study of security. So we are focusing on trust area to develop a fast scalar method to overcome the malicious attacks in nodes for security through which we can deliver more messages from source to destination without any loss in packets or data and increase the delivery rate. The result also shows the improvement in various performance parameters in the network