In this paper, a new mechanism for topology control in wireless mesh networks is proposed. We evaluate the application to this problem of the centrality metrics developed by social network analysts. Our target network is a wireless mesh network created by user hand-held devices. For this kind of networks, we aim to construct a connected dominating set that includes the most central nodes. Many advantages result from selecting just a subset of stations for routing tasks: reduction of collisions, protocol overhead, interference and energy consumption, better network organization and scalability. The resulting performance using the three most common centrality measures (degree, closeness and betweenness) is evaluated. As we are working with dynamic and decentralized networks,

Peer Reviewed