Mobility-Based Backbone Formation in Wireless Mobile Ad-hoc Networks

In this paper, the well-known network backbone formation problem is modeled as the stochastic min-degree constrained minimum spanning tree (md-MST) problem, where the link duration is associated with the edge weight. Then, a decentralized learning automata-based algorithm is proposed to form the most stable backbone of the wireless mobile ad hoc network (MANET) by finding a near optimal solution to the stochastic md-MST problem of the network topology graph. The proposed method significantly decreases the network overhead and shortens the network delay by reducing the number of intermediate forwarding hosts. It also extends the backbone lifetime by selection of the links with the maximum expected duration. The convergence of the proposed algorithm to the most stable network backbone is proven on the basis of the Martingale theorem. Several simulation experiments are conducted to investigate the efficiency of the proposed backbone formation algorithm. Numerical results show the superiority of the proposed method over the existing methods in terms of the backbone lifetime, end-to-end delay, backbone size, packet delivery ratio, and control message overhead.