Leader election on tree-based centrality in ad hoc networks

A leader in a network plays important roles for various services of the network and controls problems of the network. Therefore, when the nodes with high centrality are selected as the leader nodes in the network, the network can be more efficiently controlled, and the amount of wasted resources can be reduced. In this study, we propose a novel method for determining leader nodes by measuring centrality of nodes in the ad hoc networks. The proposed method can obtain the centrality of each node using the average depth of nodes based on the tree topology. Consequently, the amount of the information required for leader election becomes smaller and the calculation process is also simpler, compared with those of representative centrality methods in social network analysis (SNA). In the experiment, the proposed tree-based centrality technique is compared with the well-known centrality schemes such as degree, closeness, and betweenness. The proposed tree-based centrality method has similar centrality priority as degree, closeness, and betweenness and also has better performance in the comparison of the control messages generated.