Roopam Gupta’s research while affiliated with Department Of Electronics And Information Technology, India and other places

Mobile ad hoc networks are self organized, infrastructure-less networks in which each mobile host works as a router to provide connectivity within the network. Nodes out of reach to each other can communicate with the help of intermediate routers (nodes). Routing protocols are the rules which determine the way in which these routing activities are to be performed. In cluster based architecture some selected nodes (clusterheads) are identified to bear the extra burden of network activities like routing. Selection of clusterheads is a critical issue which significantly affect the performance of the network. This paper proposes an “Enhanced Performance and Trusted Weight Based Clustering (EPT-WBC) approach in which a number of performance factors such as trust, load balancing, energy consumption, mobility and battery power are considered for the selection of clusterheads. Moreover the performance of proposed scheme is compared with other existing approaches to demonstrate the effectiveness of the work.

ad hoc networks, Clustering provides a hierarchical structure in which certain nodes are assigned the extra task (such as routing) of the network. Ordinary nodes do not participate in the routing instead they rely on coordinators of the clusters (clusterheads) for packet delivery. If a suitable tap is not applied on the number of nodes that join a clusterhead as its members, formation of bottleneck can takes place at the overloaded clusterheads. The performance of the network may get affected due to the bottleneck. This paper proposes a cluster formation algorithm in which, if the number of members of a clusterhead exceeds the predefined threshold value, a procedure of cluster division is executed. This relieves the clusterheads from the burden of excessive members. Simulation study of the proposed algorithm justifies the facts by observing an improvement in the performance in terms of E2E delay, PDF and throughput. Keywordsad hoc network, clustering, clusterhead, load, energy